Merge remote-tracking branch 'origin/main' into litellm_key_update_project_assignment
Some checks failed
LiteLLM Rust DB / db-queries (push) Has been cancelled
LiteLLM Rust / rust-lint (push) Has been cancelled
LiteLLM Rust / rust-test (push) Has been cancelled
LiteLLM Rust / rust-wheel (push) Has been cancelled
Terraform Modules / fmt, validate, test (aws) (push) Has been cancelled
Terraform Modules / fmt, validate, test (gcp) (push) Has been cancelled
Terraform Provider / gofmt, vet, build, test (push) Has been cancelled
Terraform Provider / Provider endpoints vs proxy OpenAPI schema (push) Has been cancelled

# Conflicts:
#	tests/unit/proxy/management_endpoints/test_key_management_endpoints.py
This commit is contained in:
jesus 2026-10-28 01:06:17 +00:00
commit 6a2e7645aa
4797 changed files with 525332 additions and 82461 deletions

View file

@ -0,0 +1,52 @@
---
name: database-access
description: Add database queries, extract existing Prisma calls into repositories, or explicitly port queries to Rust and SQLx without expanding migration scope
---
# Database access
## Adding new database access
- Put Python database access behind an operation in `litellm/repositories`
- Give callers typed inputs and outputs, not Prisma clients, models, or query dictionaries
- Keep query execution and backend selection inside the repository
- Prisma is fine for the implementation, adding a query does not require a Rust port
- Native Rust callers use `litellm-rust/crates/db` directly
## Asked to move existing access behind a repository
- Identify the calls being moved and their transaction boundary
- Move the existing Prisma implementation into the relevant repository
- Replace those direct database calls with repository operations
- Preserve return values, errors, and transaction behavior, with focused regression tests
- Keep this an extraction, not a backend migration or unrelated cleanup
## An equivalent Rust or SQLx query already exists
- Leave it and the current backend selection unchanged unless integration is explicitly requested
- Its existence alone does not justify switching a Prisma caller to Rust
- Repository extraction can keep using Prisma even when the Rust query is available
## Asked to port a Prisma query to Rust or SQLx
- Identify the existing inputs, outputs, errors, and transaction boundary
- Check `litellm-rust/crates/db` for an equivalent query and reuse it if its behavior matches
- Otherwise follow `litellm-rust/crates/db/AGENTS.md`
- Put static SQL in `queries/` and compile-check it with SQLx
- Keep typed inputs and reusable row projections in `src/`
- Refresh SQLx query metadata after SQL changes
- Keep Python conversion in the Python bridge, not in `litellm-rust/crates/db`
- When caller integration is requested, expose the Rust operation through the bridge and call it from the repository
- A query-only port can land without switching callers
- Verify behavior against real PostgreSQL before switching callers
- Check relevant null handling, conflicts, ordering, errors, and transaction semantics
- Remove the old implementation only when no callers need it
## Migration context and boundaries
- Long-term direction is to centralize Python database access in repositories, then replace their Prisma implementations incrementally with Rust APIs
- Mixed Prisma and Rust implementations are expected during the transition, unrelated existing calls can stay as they are
- Both backends may have separate pools through PgBouncer, size them together and verify pooling-mode compatibility
- PgBouncer does not share transactions between Prisma and SQLx, keep each atomic unit of work on one backend and transaction
- Never retry a failed Rust write through Prisma when it may already have committed
- Prisma owns schema migrations today, eventual SQLx migration ownership and Prisma removal are separate planned steps, not part of a query port

View file

@ -1291,8 +1291,21 @@ jobs:
name: Install MCP SDK1 peer
command: |
uv venv --python 3.12 .venv-mcp-peer
uv pip install --python .venv-mcp-peer 'mcp==1.28.1' 'langchain-mcp-adapters==0.2.1'
uv pip install --no-config --exclude-newer "3 days" --python .venv-mcp-peer 'mcp==1.28.1' 'langchain-mcp-adapters==0.2.1'
echo "export MCP_TEST_PEER_PYTHON=$PWD/.venv-mcp-peer/bin/python" >> "$BASH_ENV"
- when:
condition:
equal: [sdk, << parameters.suite >>]
steps:
- run:
name: Install redis-server 7.2.16
command: |
mkdir -p /tmp/redis-bin
cid="$(docker create redis:7.2.16@sha256:0637954999d01b7c9ce9167db2da50656e2590d3b884f1c600c5f63bb6e6773c)"
docker cp "${cid}:/usr/local/bin/redis-server" /tmp/redis-bin/redis-server
docker rm "$cid"
/tmp/redis-bin/redis-server --version | grep -q 'v=7.2.16'
echo 'export PATH="/tmp/redis-bin:$PATH"' >> "$BASH_ENV"
- start_postgres:
image: postgres:16@sha256:e17e86066e5ef83e0952a9347f5c792b7ece00972e2aa787a6986f471b3dd3d5
server_args: "-c max_connections=300 -c shared_preload_libraries=pg_stat_statements -c pg_stat_statements.track=all -c pg_stat_statements.max=20000"
@ -1366,7 +1379,6 @@ jobs:
uv export --frozen --all-groups --all-extras --no-emit-project --no-emit-workspace --no-header --no-annotate \
| awk '/^[^ ]/ { keep = $0 ~ /^(annotated-types|coverage|pydantic|pydantic-core|typing-extensions|typing-inspection)==/ } keep' \
> /tmp/fan-in-requirements.txt
test "$(grep -c '^[^ ]' /tmp/fan-in-requirements.txt)" -eq 6
uv venv .venv --python 3.12
uv pip install --no-config --python .venv --require-hashes -r /tmp/fan-in-requirements.txt
- attach_workspace:

View file

@ -33,7 +33,7 @@ while IFS= read -r file || [ -n "$file" ]; do
*) has_backend=true ;;
esac
case "$file" in
model_prices_and_context_window.json | litellm/model_prices_and_context_window_backup.json | model_prices_and_context_window.schema.json)
model_prices_and_context_window.json | litellm/model_prices_and_context_window_backup.json)
has_cost_map=true ;;
tests/unit/proxy/*) : ;;
*) outside_cost_map_set=true ;;

View file

@ -1,90 +0,0 @@
#!/usr/bin/env bash
#
# pre-push — enforce Conventional Branches
# https://conventional-branch.github.io/
#
# Branch format: <type>/<description>
# <type> must be one of: feature, bugfix, hotfix, release, chore
#
# Protected branches (always allowed):
# - main
# - dependabot/*
# - gh-readonly-queue/*
#
# Tag pushes and branch deletions are skipped.
# Bypass: git push --no-verify.
set -eu
ZERO_OID="0000000000000000000000000000000000000000"
ZERO_OID_SHA256="0000000000000000000000000000000000000000000000000000000000000000"
ALLOWED_TYPES="feature|bugfix|hotfix|release|chore"
BRANCH_PATTERN="^(${ALLOWED_TYPES})/.+"
PROTECTED_NAMES="main"
PROTECTED_PREFIXES="dependabot/ gh-readonly-queue/"
is_protected() {
branch="$1"
for name in $PROTECTED_NAMES; do
if [ "$branch" = "$name" ]; then
return 0
fi
done
for prefix in $PROTECTED_PREFIXES; do
case "$branch" in "$prefix"*) return 0 ;; esac
done
return 1
}
invalid=""
while read -r local_ref local_oid remote_ref remote_oid; do
# Branch deletion (no local commit being pushed).
if [ "$local_oid" = "$ZERO_OID" ] || [ "$local_oid" = "$ZERO_OID_SHA256" ]; then
continue
fi
# Only validate branch pushes; ignore tags and other ref namespaces.
case "$remote_ref" in
refs/heads/*) ;;
*) continue ;;
esac
branch="${remote_ref#refs/heads/}"
if is_protected "$branch"; then
continue
fi
if ! printf '%s' "$branch" | grep -Eq "$BRANCH_PATTERN"; then
invalid="$invalid $branch"
fi
done
if [ -n "$invalid" ]; then
cat >&2 <<EOF
✗ Branch name does not follow Conventional Branches.
Invalid:$invalid
Expected: <type>/<description>
Allowed types: feature, bugfix, hotfix, release, chore
Examples:
feature/weighted-round-robin
bugfix/streaming-empty-chunks
chore/bump-deps
hotfix/auth-bypass
Protected (always allowed): main, dependabot/*, gh-readonly-queue/*.
See https://conventional-branch.github.io/
Rename with: git branch -m <new-name>
To bypass (use sparingly): git push --no-verify
EOF
exit 1
fi
exit 0

6
.github/CODEOWNERS vendored
View file

@ -1,2 +1,4 @@
/model_prices_and_context_window.json @mateo-berri @ryan-crabbe-berri @kerry-berri
/litellm/model_prices_and_context_window_backup.json @mateo-berri @ryan-crabbe-berri @kerry-berri
/model_prices_and_context_window.json @mateo-berri @ryan-crabbe-berri @kerry-berri @joshua-berri
/litellm/model_prices_and_context_window_backup.json @mateo-berri @ryan-crabbe-berri @kerry-berri @joshua-berri
/helm/ @yassin-berriai
/terraform/ @yassin-berriai

View file

@ -4,6 +4,13 @@
Drop every section you have nothing to put in, heading included: a bare "## Relevant issues" or
"## Affected release" with nothing under it must not appear in the final description -->
<!-- Two or three plain sentences a teammate can read in 10 seconds, written the way you'd explain it in Slack
Say what the PR does for users and why, then anything that keeps it working over time (a sync job, a config someone owns)
No headings, bullets, tables, or code names here. Everything else goes inside the Details block below -->
<details>
<summary>Details</summary>
## TLDR
<!-- Fill in the bullets below and keep each one short and concrete: one line per bullet, roughly 10 words max
@ -164,3 +171,5 @@ Example checklists:
## Final Attestation
- [ ] The tests check the right things, including the edge cases, and regressions in the respective real-world customer use-cases are not possible after this PR
</details>

View file

@ -5,7 +5,7 @@ description: >-
test_paths:
- reason: >-
litellm.agent() end-to-end suite. It drives the real claude, codex and opencode CLIs and
litellm.agent() end-to-end suite. It drives the real claude, codex, opencode and pi CLIs and
deepagents against a live LiteLLM AI Gateway, so it needs those binaries on PATH plus
LITELLM_PROXY_API_BASE / LITELLM_PROXY_API_KEY, and skips without them. Run manually
before changing litellm/harness; the mocked coverage runs in tests/unit/harness and

View file

@ -1,69 +0,0 @@
import os
import re
import sys
import xml.etree.ElementTree as ET
from pathlib import Path
from typing import Final
sys.path.insert(0, str(Path(__file__).resolve().parents[2] / "tests/e2e"))
from coverage_registry.management_cases import MANAGEMENT_CASES
def main() -> int:
selected: Final = tuple(sys.argv[2:])
try:
report: Final = ET.parse(Path(sys.argv[1])).getroot()
except (ET.ParseError, OSError):
_ = sys.stdout.write("::error::could not read the test execution report\n")
return 1
cases: Final = tuple(report.iter("testcase"))
expected_count: Final = os.environ.get("E2E_REQUIRED_TEST_COUNT")
passed: Final = frozenset(
case.get("file") for case in cases if all(case.find(tag) is None for tag in ("skipped", "failure", "error"))
)
missing: Final = tuple(path for path in selected if path not in passed)
required_nodes: Final = frozenset(case.node for case in MANAGEMENT_CASES if case.node.split("::", 1)[0] in selected)
passed_nodes: Final = frozenset(
prop.get("value")
for case in cases
if all(case.find(tag) is None for tag in ("skipped", "failure", "error"))
for prop in case.findall("./properties/property")
if prop.get("name") == "management_node"
)
missing_nodes: Final = required_nodes - passed_nodes
for node in sorted(missing_nodes):
_ = sys.stdout.write(f"::error::required management case did not pass: {node}\n")
for path in selected:
collected: Final = sum(case.get("file") == path for case in cases)
skipped: Final = sum(case.get("file") == path and case.find("skipped") is not None for case in cases)
_ = sys.stdout.write(f"{path}: {collected} collected, {skipped} skipped\n")
for case in cases:
if case.get("file") != path or all(case.find(tag) is None for tag in ("failure", "error", "skipped")):
continue
outcome = "skipped" if case.find("skipped") is not None else "failed"
_ = sys.stdout.write(f" {outcome}: {case.get('classname', '')}::{case.get('name', '')}\n")
for prop in case.findall("./properties/property"):
name = prop.get("name", "")
value = prop.get("value", "")
if name in ("oauth_failure_phase", "oauth_exception_type", "oauth_frame") and re.fullmatch(
r"[A-Za-z0-9_.:<>-]{1,240}", value
):
_ = sys.stdout.write(f" {name}: {value}\n")
if expected_count is not None and (
len(cases) != int(expected_count) or any(case.find("skipped") is not None for case in cases)
):
_ = sys.stdout.write("::error::required test count was not met or a required case was skipped\n")
return 1
if (
selected
and not missing
and not missing_nodes
and not any(case.find(tag) is not None for case in cases for tag in ("failure", "error"))
):
return 0
_ = sys.stdout.write("::error::every selected file must execute a passing test, with no failures or errors\n")
return 1
if __name__ == "__main__":
sys.exit(main())

View file

@ -32,6 +32,7 @@ IGNORE_ARG_RE: Final = re.compile(r"--ignore(?:-glob)?[= ](\S+)")
DOCKERFILE_TOKEN_RE = re.compile(r"[A-Za-z0-9_./-]*Dockerfile[A-Za-z0-9_.-]*")
COMMENT_RE = re.compile(r"^\s*#.*$", re.MULTILINE)
GLOB_CHARS = frozenset("*?")
DEDICATED_INTEGRATION_WORKFLOWS: Final = MappingProxyType({"image-scan.yml": "tests/integration/database/image/"})
# Trees whose jobs are sharded with no catch-all bucket, so every child that holds
# tests has to be named by some shard or it runs nowhere. A child listed here is
@ -560,6 +561,17 @@ def _integration_github_files(runner: pathlib.Path) -> frozenset[str]:
return frozenset(ast.literal_eval(values))
def _owns(workflow: str, relative_path: str) -> bool:
owned_root: Final = DEDICATED_INTEGRATION_WORKFLOWS.get(workflow)
return owned_root is not None and relative_path.startswith(owned_root)
def _tokens_outside_owner(relative_path: str, workflow_tokens: Mapping[str, frozenset[str]]) -> frozenset[str]:
return frozenset[str]().union(
*(tokens for workflow, tokens in workflow_tokens.items() if not _owns(workflow, relative_path))
)
def _integration_ownership(repo_root: pathlib.Path = REPO_ROOT) -> tuple[frozenset[str], tuple[Finding, ...]]:
runner: Final = repo_root / "tests/integration/run.py"
if not runner.exists():
@ -611,15 +623,20 @@ def _integration_ownership(repo_root: pathlib.Path = REPO_ROOT) -> tuple[frozens
)
!= 1
)
gha_tokens: Final = _invoked_test_tokens(
scalar
for path in (repo_root / ".github/workflows").glob("*.y*ml")
for scalar in _scalars(yaml.safe_load(path.read_text()), path.name)
workflow_tokens: Final = MappingProxyType(
{
path.name: _invoked_test_tokens(_scalars(yaml.safe_load(path.read_text()), path.name))
for path in (repo_root / ".github/workflows").glob("*.y*ml")
}
)
gha_tokens: Final = frozenset[str]().union(*workflow_tokens.values())
circle_owned: Final = (
frozenset(str(path.relative_to(repo_root)) for path in integration_root.rglob("test_*.py")) - github_files
)
findings: Final = (
tuple(
Finding(path, "integration contract is also selected by GitHub Actions")
for path in paths
Finding(path, "integration file outside GITHUB_FILES is selected by GitHub Actions")
for path in sorted(circle_owned)
if any(_token_covers(token, path) for token in gha_tokens)
)
+ tuple(
@ -632,6 +649,11 @@ def _integration_ownership(repo_root: pathlib.Path = REPO_ROOT) -> tuple[frozens
for path in sorted(github_files)
if not (repo_root / path).is_file()
)
+ tuple(
Finding(path, "GitHub-owned integration contract runs outside a dedicated integration workflow")
for path in sorted(github_files)
if any(_token_covers(token, path) for token in _tokens_outside_owner(path, workflow_tokens))
)
)
browser_commands: Final = tuple(
scalar.value
@ -688,13 +710,9 @@ def main() -> int:
scalars = _all_scalars()
integration_paths, ownership_findings = _integration_ownership()
test_findings = (
_uncovered_tests(
allowlist,
_invoked_selections(scalars)
+ (Selection(included=integration_paths, ignored=frozenset()),),
)
+ ownership_findings
test_findings = _uncovered_tests(
allowlist,
_invoked_selections(scalars) + (Selection(included=integration_paths, ignored=frozenset()),),
)
dockerfile_findings = _uncovered_dockerfiles(allowlist, _built_dockerfile_tokens(scalars))
stale_findings = _stale_allowlist_paths(allowlist, test_files=_test_files(), dockerfiles=_dockerfiles())
@ -711,13 +729,25 @@ def main() -> int:
test_findings,
"Add each to a job's test path, or list it in .github/ci-coverage-allowlist.yml with a reason.",
)
if ownership_findings:
browser_remedy = (
" CircleCI's browser group owns the integrationCritical Playwright contracts, so no GitHub workflow may "
"select them and the default Playwright config must exclude them."
if any("browser" in finding.detail or "Playwright" in finding.detail for finding in ownership_findings)
else ""
)
_report(
"integration contracts with the wrong CI owner",
ownership_findings,
"Follow tests/integration/AGENTS.md, which says which CI job owns each integration file." + browser_remedy,
)
if dockerfile_findings:
_report(
"Dockerfiles that no CI job builds",
dockerfile_findings,
"Build each in a workflow, or list it in .github/ci-coverage-allowlist.yml with a reason.",
)
if stale_findings or test_findings or dockerfile_findings:
if stale_findings or test_findings or ownership_findings or dockerfile_findings:
return 1
_write(

View file

@ -1,319 +0,0 @@
import asyncio
import aiohttp
import json
import math
from typing import Any
# Asynchronously fetch data from a given URL
async def fetch_data(url):
try:
# Create an asynchronous session
async with aiohttp.ClientSession() as session:
# Send a GET request to the URL
async with session.get(url) as resp:
# Raise an error if the response status is not OK
resp.raise_for_status()
# Parse the response JSON
resp_json = await resp.json()
print("Fetch the data from URL.")
# Return the 'data' field from the JSON response
return resp_json['data']
except Exception as e:
# Print an error message if fetching data fails
print("Error fetching data from URL:", e)
return None
FRIENDLI_API_URL = "https://api.friendli.ai/serverless/v1/models"
FRIENDLI_PROVIDER = "friendliai"
INHERITABLE_BASE_KEYS = (
"supports_pdf_input",
"supports_assistant_prefill",
"supports_adaptive_thinking",
"supports_output_config",
)
REASONING_EFFORT_LEVEL_ORDER = ("none", "minimal", "low", "medium", "high", "xhigh", "max")
def _find_base_model_entry(base_model: str, local_data: dict) -> str | None:
if not base_model:
return None
bm_tail = base_model.split("/")[-1].lower()
if base_model in local_data:
return base_model
for key in local_data:
if key.startswith("sample_spec") or key == "fallback_generalizations":
continue
if key.split("/")[-1].lower() == bm_tail:
return key
return None
def _reasoning_effort_levels(reasoning_options: list) -> list:
offered = {
val
for opt in reasoning_options or []
if opt.get("type") == "effort"
for val in opt.get("values", [])
}
return [level for level in REASONING_EFFORT_LEVEL_ORDER if level in offered]
def _valid_token_price(value: object) -> bool:
try:
price = float(value) # pyright: ignore[reportArgumentType] # non-numeric values are rejected via the except
except (TypeError, ValueError):
return False
return math.isfinite(price) and price >= 0
def _has_valid_token_prices(pricing: dict | None) -> bool:
prices = pricing or {}
return _valid_token_price(prices.get("input")) and _valid_token_price(prices.get("output"))
def _pricing(pricing: dict) -> dict:
out: dict[str, Any] = {}
if not pricing:
return out
if "input" in pricing:
out["input_cost_per_token"] = float(pricing["input"])
if "output" in pricing:
out["output_cost_per_token"] = float(pricing["output"])
if "input_cache_read" in pricing and pricing["input_cache_read"] is not None:
out["cache_read_input_token_cost"] = float(pricing["input_cache_read"])
return out
def _modality_flags(input_mods: list) -> dict:
mods = input_mods or []
has_image = "image" in mods
return {
"supports_vision": has_image,
"supports_image_input": has_image,
"supports_video_input": "video" in mods,
}
def transform_friendli_data(data: list, local_data: dict) -> dict:
transformed: dict[str, dict] = {}
if not data:
return transformed
for model in data:
# An unpriced row must never wholesale-replace an already priced local entry:
# missing prices cost-calculate as zero, silently zeroing tracked spend
if not _has_valid_token_prices(model.get("pricing")):
continue
model_id = model["id"]
base_model = model.get("base_model") or ""
entry: dict[str, Any] = {
"litellm_provider": FRIENDLI_PROVIDER,
}
base_key = _find_base_model_entry(base_model, local_data)
if base_key:
base_entry = local_data[base_key]
for k in INHERITABLE_BASE_KEYS:
if k in base_entry:
entry[k] = base_entry[k]
ctx = model.get("context_length")
if ctx is not None:
entry["max_input_tokens"] = int(ctx)
max_out = model.get("max_completion_tokens")
if max_out is not None:
entry["max_output_tokens"] = int(max_out)
entry["max_tokens"] = int(max_out)
pricing = _pricing(model.get("pricing", {}))
entry.update(pricing)
entry["supports_prompt_caching"] = "cache_read_input_token_cost" in pricing
reasoning = model.get("reasoning") is True
entry["supports_reasoning"] = reasoning
if reasoning:
entry["reasoning_effort_levels"] = _reasoning_effort_levels(
model.get("reasoning_options", [])
)
func = model.get("functionality", {})
entry["supports_function_calling"] = func.get("tool_call") is True
entry["supports_parallel_function_calling"] = func.get("parallel_tool_call") is True
is_struct = func.get("structured_output") is True
entry["supports_response_schema"] = is_struct
entry["supports_native_structured_output"] = is_struct
entry["supports_system_messages"] = func.get("system_messages") is True
entry["supports_tool_choice"] = func.get("tool_choice") is True
entry.update(_modality_flags(model.get("input_modalities", [])))
entry["mode"] = model.get("mode", "chat")
desc = model.get("description")
if desc:
entry["comment"] = desc
dep = model.get("deprecation_date")
if dep:
entry["deprecation_date"] = dep.split("T")[0]
entry["source"] = FRIENDLI_API_URL
transformed[f"{FRIENDLI_PROVIDER}/{model_id}"] = entry
return transformed
# Synchronize local data with remote data
def sync_local_data_with_remote(local_data, remote_data, replace_keys=frozenset()):
# Update existing keys in local_data with values from remote_data
# (replace_keys entries are swapped wholesale so a field the remote catalog
# dropped, e.g. cache pricing, cannot survive as a stale value)
for key in (set(local_data) & set(remote_data)):
if key in replace_keys:
local_data[key] = remote_data[key]
else:
local_data[key].update(remote_data[key])
# Add new keys from remote_data to local_data
for key in (set(remote_data) - set(local_data)):
local_data[key] = remote_data[key]
# Write data to the json file
def write_to_file(file_path, data):
try:
# Open the file in write mode
with open(file_path, "w") as file:
# Dump the data as JSON into the file
json.dump(data, file, indent=4)
print("Values updated successfully.")
except Exception as e:
# Print an error message if writing to file fails
print("Error updating JSON file:", e)
# Update the existing models and add the missing models for OpenRouter
def transform_openrouter_data(data):
transformed = {}
if not data:
return transformed
for row in data:
# Add the fields 'max_tokens' and 'input_cost_per_token'
obj = {
"max_tokens": row["context_length"],
"input_cost_per_token": float(row["pricing"]["prompt"]),
}
# Add 'max_output_tokens' as a field if it is not None
if "top_provider" in row and "max_completion_tokens" in row["top_provider"] and row["top_provider"]["max_completion_tokens"] is not None:
obj['max_output_tokens'] = int(row["top_provider"]["max_completion_tokens"])
# Add the field 'output_cost_per_token'
obj.update({
"output_cost_per_token": float(row["pricing"]["completion"]),
})
# Add field 'input_cost_per_image' if it exists and is non-zero
if "pricing" in row and "image" in row["pricing"] and float(row["pricing"]["image"]) != 0.0:
obj['input_cost_per_image'] = float(row["pricing"]["image"])
# Add the fields 'litellm_provider' and 'mode'
obj.update({
"litellm_provider": "openrouter",
"mode": "chat"
})
# Add the 'supports_vision' field if the modality is 'multimodal'
if row.get('architecture', {}).get('modality') == 'multimodal':
obj['supports_vision'] = True
# Use a composite key to store the transformed object
transformed[f'openrouter/{row["id"]}'] = obj
return transformed
# Update the existing models and add the missing models for Vercel AI Gateway
def transform_vercel_ai_gateway_data(data):
transformed = {}
if not data:
return transformed
for row in data:
# Rows without token pricing or token limits (video/embedding models) previously KeyError'd the whole sync
if any(row.get(k) is None for k in ("context_window", "max_tokens")) or any(
row.get("pricing", {}).get(k) is None for k in ("input", "output")
):
continue
obj = {
"max_tokens": row["context_window"],
"input_cost_per_token": float(row["pricing"]["input"]),
"output_cost_per_token": float(row["pricing"]["output"]),
'max_output_tokens': row['max_tokens'],
'max_input_tokens': row["context_window"],
}
# Handle cache pricing if available
if "pricing" in row:
if "input_cache_read" in row["pricing"] and row["pricing"]["input_cache_read"] is not None:
obj['cache_read_input_token_cost'] = float(f"{float(row['pricing']['input_cache_read']):e}")
if "input_cache_write" in row["pricing"] and row["pricing"]["input_cache_write"] is not None:
obj['cache_creation_input_token_cost'] = float(f"{float(row['pricing']['input_cache_write']):e}")
mode = "embedding" if "embedding" in row["id"].lower() else "chat"
obj.update({"litellm_provider": "vercel_ai_gateway", "mode": mode})
transformed[f'vercel_ai_gateway/{row["id"]}'] = obj
return transformed
# Load local data from a specified file
def load_local_data(file_path):
try:
# Open the file in read mode
with open(file_path, "r") as file:
# Load and return the JSON data
return json.load(file)
except FileNotFoundError:
# Print an error message if the file is not found
print("File not found:", file_path)
return None
except json.JSONDecodeError as e:
# Print an error message if JSON decoding fails
print("Error decoding JSON:", e)
return None
def main():
local_file_path = "model_prices_and_context_window.json" # Path to the local data file
openrouter_url = "https://openrouter.ai/api/v1/models" # URL to fetch OpenRouter data
vercel_ai_gateway_url = "https://ai-gateway.vercel.sh/v1/models" # URL to fetch Vercel AI Gateway data
# Load local data from file
local_data = load_local_data(local_file_path)
# Fetch OpenRouter data
openrouter_data = asyncio.run(fetch_data(openrouter_url))
# Transform the fetched OpenRouter data
openrouter_data = transform_openrouter_data(openrouter_data)
# Fetch Vercel AI Gateway data
vercel_data = asyncio.run(fetch_data(vercel_ai_gateway_url))
# Transform the fetched Vercel AI Gateway data
vercel_data = transform_vercel_ai_gateway_data(vercel_data)
friendli_data = asyncio.run(fetch_data(FRIENDLI_API_URL))
friendli_data = transform_friendli_data(friendli_data, local_data)
# Combine both datasets
all_remote_data = {**openrouter_data, **vercel_data, **friendli_data}
# If both local and openrouter data are available, synchronize and save
if local_data and all_remote_data:
sync_local_data_with_remote(local_data, all_remote_data, replace_keys=frozenset(friendli_data))
write_to_file(local_file_path, local_data)
else:
print("Failed to fetch model data from either local file or URL.")
# Entry point of the script
if __name__ == "__main__":
main()

View file

@ -1,39 +0,0 @@
name: Updates model_prices_and_context_window.json and Create Pull Request
on:
schedule:
- cron: "0 0 * * 0" # Run every Sundays at midnight
#- cron: "0 0 * * *" # Run daily at midnight
permissions:
contents: write
pull-requests: write
jobs:
auto_update_price_and_context_window:
if: github.repository == 'BerriAI/litellm'
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:
persist-credentials: false
- name: Set up uv
uses: ./.github/actions/setup-uv-with-retries
with:
version: "0.10.9"
- name: Update JSON Data
run: |
uv run --frozen --with 'aiohttp==3.13.3' python ".github/scripts/auto_update_price_and_context_window_file.py"
- name: Regenerate JSON Schema
run: |
uv run --frozen python ci_cd/generate_model_prices_schema.py
- name: Create Pull Request
run: |
git add model_prices_and_context_window.json model_prices_and_context_window.schema.json
git commit -m "Update model_prices_and_context_window.json file: $(date +'%Y-%m-%d')"
gh pr create --title "Update model_prices_and_context_window.json file" \
--body "Automated update for model_prices_and_context_window.json" \
--head auto-update-price-and-context-window-$(date +'%Y-%m-%d') \
--base main
env:
GH_TOKEN: ${{ secrets.GH_TOKEN }}

View file

@ -40,4 +40,4 @@ jobs:
HEAD_SHA: ${{ github.event.pull_request.head.sha }}
HEAD_REF: ${{ github.event.pull_request.head.ref }}
run: |
uv run --frozen python ci_cd/cost_map_guard.py --base "$MERGE_BASE" --head "$HEAD_SHA" --head-ref "$HEAD_REF"
uv run --no-project python ci_cd/cost_map_guard.py --base "$MERGE_BASE" --head "$HEAD_SHA" --head-ref "$HEAD_REF"

View file

@ -0,0 +1,91 @@
name: Interactions spec canary
on:
schedule:
- cron: "17 7 * * *"
pull_request:
paths:
- .github/workflows/interactions-spec-canary.yml
- tests/unit/interactions/**
- litellm/types/interactions/**
workflow_dispatch:
permissions: {}
concurrency: interactions-spec-canary
jobs:
interactions-spec-canary:
name: interactions-spec-canary
if: github.repository == 'BerriAI/litellm'
runs-on: ubuntu-latest
timeout-minutes: 15
permissions:
contents: read
outputs:
compliance: ${{ steps.compliance.outcome }}
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:
persist-credentials: false
- name: Set up Python
uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0
with:
python-version: "3.12"
- name: Set up uv
uses: ./.github/actions/setup-uv-with-retries
with:
version: "0.10.9"
- name: Install dependencies
run: .github/scripts/uv_sync_with_retries.sh --frozen --group ci --group proxy-dev --extra proxy --no-install-project
- name: Run the compliance tests against Google's live spec
id: compliance
run: |
uv run --no-sync pytest tests/unit/interactions/test_openapi_compliance.py \
--interactions-spec-url=https://ai.google.dev/static/api/interactions.openapi.json \
--tb=short -vv -p no:rerunfailures -p no:pytest-retry \
--allow-hosts=127.0.0.1,::1,localhost,ai.google.dev
tracking-issue:
name: tracking-issue
needs: interactions-spec-canary
if: always() && github.event_name != 'pull_request' && github.ref_name == github.event.repository.default_branch && github.repository == 'BerriAI/litellm'
runs-on: ubuntu-latest
timeout-minutes: 5
permissions:
issues: write
steps:
- name: File the tracking issue unless the compliance step passed, close it when it did
uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1
env:
COMPLIANCE_OUTCOME: ${{ needs.interactions-spec-canary.outputs.compliance }}
RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
with:
script: |
const title = "Interactions spec canary is red";
const outcome = process.env.COMPLIANCE_OUTCOME;
const runUrl = process.env.RUN_URL;
const query = `repo:${context.repo.owner}/${context.repo.repo} is:issue is:open in:title "${title}"`;
const { data: found } = await github.rest.search.issuesAndPullRequests({ q: query, per_page: 20 });
const open = found.items.find((issue) => issue.title === title);
if (outcome === "success") {
if (open === undefined) return;
await github.rest.issues.createComment({ ...context.repo, issue_number: open.number, body: `Green again: ${runUrl}` });
await github.rest.issues.update({ ...context.repo, issue_number: open.number, state: "closed" });
return;
}
const state = outcome === "failure" ? "red" : `not run (compliance step outcome: ${outcome || "none"})`;
if (open !== undefined) {
await github.rest.issues.createComment({ ...context.repo, issue_number: open.number, body: `Still ${state}: ${runUrl}` });
return;
}
await github.rest.issues.create({
...context.repo,
title,
assignees: ["mateo-berri"],
body: `The scheduled run of tests/unit/interactions/test_openapi_compliance.py against Google's live Interactions spec is ${state}: ${runUrl}\n\nlitellm/types/interactions/README.md says how to refresh the snapshot and regenerate the types.`,
});

49
.github/workflows/spend-logs-guard.yml vendored Normal file
View file

@ -0,0 +1,49 @@
name: Spend logs guard
on: # zizmor: ignore[dangerous-triggers] runs the base branch's code only; the PR's diff is read as data and never executed
pull_request_target:
branches:
- main
pull_request_review:
types:
- submitted
- dismissed
permissions:
contents: read
pull-requests: read
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number }}
cancel-in-progress: true
jobs:
spend-logs-guard:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:
ref: ${{ github.event.pull_request.base.ref }}
persist-credentials: false
- name: Fetch the pull request head, its merge base and its reviews
id: revisions
env:
GH_TOKEN: ${{ github.token }}
BASE_SHA: ${{ github.event.pull_request.base.sha }}
HEAD_SHA: ${{ github.event.pull_request.head.sha }}
PR_NUMBER: ${{ github.event.pull_request.number }}
run: |
merge_base="$(gh api "repos/${GITHUB_REPOSITORY}/compare/${BASE_SHA}...${HEAD_SHA}" --jq '.merge_base_commit.sha')"
git fetch --no-tags --depth=1 origin "$merge_base" "$HEAD_SHA"
gh api "repos/${GITHUB_REPOSITORY}/pulls/${PR_NUMBER}/reviews" --paginate --slurp | jq 'add // []' > "$RUNNER_TEMP/reviews.json"
echo "merge_base=$merge_base" >> "$GITHUB_OUTPUT"
- name: Set up uv
uses: ./.github/actions/setup-uv-with-retries
with:
version: "0.10.9"
- name: Run the guard
env:
MERGE_BASE: ${{ steps.revisions.outputs.merge_base }}
HEAD_SHA: ${{ github.event.pull_request.head.sha }}
run: |
uv run --no-project --with pydantic==2.14.1 python ci_cd/spend_logs_guard.py --base "$MERGE_BASE" --head "$HEAD_SHA" --approver kerry-berri --reviews "$RUNNER_TEMP/reviews.json"

View file

@ -1,70 +0,0 @@
name: Sync Together AI model registry
on:
schedule:
- cron: "30 6 * * *"
workflow_dispatch:
permissions:
contents: write
pull-requests: write
jobs:
sync_together_ai_models:
if: github.repository == 'BerriAI/litellm'
runs-on: ubuntu-latest
env:
BASE_BRANCH: ${{ github.event.repository.default_branch }}
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:
ref: ${{ env.BASE_BRANCH }}
persist-credentials: false
- name: Set up uv
uses: ./.github/actions/setup-uv-with-retries
with:
version: "0.10.9"
- name: Look for an already-open sync PR
id: existing
run: |
open_pr="$(gh pr list --repo "$GITHUB_REPOSITORY" --state open --limit 1000 --json headRefName \
--jq '[.[].headRefName | select(startswith("litellm_together_registry_sync_"))] | first // empty')"
echo "open_pr=$open_pr" >> "$GITHUB_OUTPUT"
if [ -n "$open_pr" ]; then
echo "An open sync PR already exists on branch $open_pr; skipping this run."
fi
env:
GH_TOKEN: ${{ secrets.GH_TOKEN || github.token }}
- name: Run the sync
if: steps.existing.outputs.open_pr == ''
run: |
uv run --frozen python scripts/sync_together_ai_models.py --write --pr-body-file "$RUNNER_TEMP/pr_body.md"
env:
TOGETHER_API_KEY: ${{ secrets.TOGETHER_API_KEY }}
- name: Regenerate the JSON schema
if: steps.existing.outputs.open_pr == ''
run: |
uv run --frozen python ci_cd/generate_model_prices_schema.py
- name: Create a pull request when the registry changed
if: steps.existing.outputs.open_pr == ''
run: |
if git diff --quiet; then
echo "Registry already in sync; no PR needed."
exit 0
fi
branch="litellm_together_registry_sync_$(date +'%Y-%m-%d')"
git config user.name "github-actions[bot]"
git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
git checkout -b "$branch"
git add model_prices_and_context_window.json \
litellm/model_prices_and_context_window_backup.json \
model_prices_and_context_window.schema.json
git commit -m "feat(models): sync together_ai model registry $(date +'%Y-%m-%d')"
gh auth setup-git
git push origin "$branch"
gh pr create --title "feat(models): sync together_ai model registry" \
--body-file "$RUNNER_TEMP/pr_body.md" \
--head "$branch" \
--base "$BASE_BRANCH"
env:
GH_TOKEN: ${{ secrets.GH_TOKEN || github.token }}

View file

@ -21,7 +21,7 @@ jobs:
pull-requests: read
actions: read
runs-on: ubuntu-latest
timeout-minutes: 15
timeout-minutes: 20
services:
redis:
image: redis@sha256:e2debfb7956fa12c7ddc79d7e645c8cf26b30c99a6e9161ea9bf4171e1668a5f
@ -524,6 +524,12 @@ jobs:
- name: check_provider_folders_documented
run: uv run --no-sync python ./tests/code_coverage_tests/check_provider_folders_documented.py
- name: check_provider_endpoints_support_generated
run: uv run --no-sync python scripts/generate_provider_endpoints_support.py --check
- name: test_generate_provider_endpoints_support
run: uv run --no-sync pytest -q --noconftest -p no:cacheprovider -c /dev/null tests/code_coverage_tests/test_generate_provider_endpoints_support.py
- name: check_prisma_binary_cache
run: uv run --no-sync python ./tests/code_coverage_tests/check_prisma_binary_cache.py
@ -539,8 +545,8 @@ jobs:
- name: test_unit_passed_gate
run: uv run --no-sync pytest -q --noconftest -p no:cacheprovider -c /dev/null tests/code_coverage_tests/test_unit_passed_gate.py
- name: test_e2e_changed_gate
run: uv run --no-sync pytest -q --noconftest -p no:cacheprovider -c /dev/null tests/code_coverage_tests/test_e2e_changed_gate.py tests/code_coverage_tests/test_e2e_idp_stack.py
- name: test_e2e_idp_stack
run: uv run --no-sync pytest -q --noconftest -p no:cacheprovider -c /dev/null tests/code_coverage_tests/test_e2e_idp_stack.py
- name: test_e2e_metadata
env:

View file

@ -45,7 +45,7 @@ jobs:
strategy:
fail-fast: false
matrix:
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14", "3.15"]
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
@ -132,7 +132,7 @@ jobs:
strategy:
fail-fast: false
matrix:
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14", "3.15"]
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:

View file

@ -1,180 +0,0 @@
name: MCP OAuth happy path
on:
pull_request:
paths:
- '.github/workflows/test-mcp-oauth-e2e.yml'
- '.github/e2e-stack/**'
- 'tests/e2e/*.py'
- 'tests/e2e/pytest.ini'
- 'tests/e2e/idp_realm.json'
- 'tests/e2e/mcp/**'
- 'litellm/experimental_mcp_client/**'
- 'litellm/proxy/_experimental/mcp_server/**'
- 'litellm/proxy/auth/**'
- 'litellm/proxy/management_endpoints/mcp_management_endpoints.py'
- 'litellm/proxy/_types.py'
- 'litellm/types/mcp_server/mcp_server_manager.py'
- 'litellm/proxy/management_endpoints/*sso*.py'
- 'litellm/proxy/management_endpoints/sso/**'
- 'litellm/proxy/common_utils/encrypt_decrypt_utils.py'
- 'litellm/proxy/proxy_server.py'
- 'litellm/proxy/schema.prisma'
- 'ui/litellm-dashboard/src/app/connect/**'
- 'ui/litellm-dashboard/src/app/mcp/oauth/**'
- 'pyproject.toml'
- 'uv.lock'
workflow_dispatch:
permissions: {}
concurrency:
group: mcp-oauth-${{ github.ref }}
cancel-in-progress: true
jobs:
oauth:
if: github.event_name == 'workflow_dispatch' || github.event.pull_request.head.repo.full_name == github.repository
runs-on: ubuntu-latest
environment: e2e-changed
timeout-minutes: 45
permissions:
contents: read
id-token: write
services:
postgres:
image: postgres:16.6
env:
POSTGRES_USER: litellm
POSTGRES_PASSWORD: dbpassword9090
POSTGRES_DB: litellm
ports:
- 5432:5432
options: >-
--health-cmd "pg_isready -U litellm"
--health-interval 5s
--health-timeout 5s
--health-retries 10
env:
DATABASE_HOST: 127.0.0.1
DATABASE_PORT: '5432'
DATABASE_USER: litellm
DATABASE_PASSWORD: dbpassword9090
DATABASE_NAME: litellm
DATABASE_URL: postgresql://litellm:dbpassword9090@127.0.0.1:5432/litellm
E2E_KEYCLOAK_URL: http://127.0.0.1:8081
E2E_KEYCLOAK_ADMIN_USER: admin
E2E_KEYCLOAK_ADMIN_PASSWORD: e2e-ephemeral-idp-not-a-secret
E2E_FIXTURE_MODE: live
E2E_PROVIDER_CACHE: '0'
E2E_MCP_OAUTH_LIVE: '1'
E2E_REQUIRED_TEST_COUNT: '4'
steps:
- name: Checkout the tested source
uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:
ref: ${{ github.event.pull_request.head.sha || github.sha }}
persist-credentials: false
- name: Require and materialize the upstream login
env:
STORAGE_STATE: ${{ secrets.E2E_LINEAR_STORAGE_STATE_B64 }}
run: |
umask 077
python3 - <<'PY'
import base64
import json
import os
import secrets
from pathlib import Path
encoded = os.environ.get("STORAGE_STATE", "")
if not encoded:
raise SystemExit("E2E_LINEAR_STORAGE_STATE_B64 is required; capture and provision a test-account login")
state = json.loads(base64.b64decode(encoded, validate=True))
if not isinstance(state, dict) or not state.get("cookies"):
raise SystemExit("The captured login must contain browser cookies")
directory = Path(os.environ["RUNNER_TEMP"]) / "mcp-oauth-private"
directory.mkdir(mode=0o700)
path = directory / "linear-state.json"
path.write_text(json.dumps(state))
with open(os.environ["GITHUB_ENV"], "a") as output:
output.write(f"E2E_LINEAR_STORAGE_STATE={path}\n")
for name in ("LITELLM_MASTER_KEY", "LITELLM_SALT_KEY"):
value = "sk-e2e-" + secrets.token_hex(24)
print(f"::add-mask::{value}")
output.write(f"{name}={value}\n")
PY
- uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0
with:
python-version: '3.13'
- uses: ./.github/actions/setup-uv-with-retries
with:
version: '0.10.9'
- uses: ./.github/actions/cache-cargo-build
- name: Install the frozen E2E environment
run: |
.github/scripts/uv_sync_with_retries.sh --frozen --extra proxy --extra proxy-runtime --extra extra_proxy --group ci --group proxy-dev --group e2e-dev
uv run --no-sync python scripts/prisma_generate_if_needed.py
uv run --no-sync playwright install --with-deps chromium
- name: Configure license access
id: aws
uses: aws-actions/configure-aws-credentials@e7f100cf4c008499ea8adda475de1042d6975c7b # v6.2.0
with:
role-to-assume: ${{ vars.E2E_AWS_ROLE_TO_ASSUME }}
aws-region: us-east-1
role-session-name: mcp-oauth-${{ github.run_id }}
role-duration-seconds: 900
output-env-credentials: false
output-credentials: true
- name: Load the E2E license
env:
AWS_ACCESS_KEY_ID: ${{ steps.aws.outputs.aws-access-key-id }}
AWS_SECRET_ACCESS_KEY: ${{ steps.aws.outputs.aws-secret-access-key }}
AWS_SESSION_TOKEN: ${{ steps.aws.outputs.aws-session-token }}
AWS_DEFAULT_REGION: us-east-1
run: |
license="$(aws secretsmanager get-secret-value --secret-id litellm-e2e-changed-license --query SecretString --output text)"
test -n "${license}"
echo "::add-mask::${license}"
echo "LITELLM_LICENSE=${license}" >> "${GITHUB_ENV}"
- uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4.4.0
with:
node-version-file: ui/litellm-dashboard/.nvmrc
- name: Build the gateway consent UI at the tested commit
run: |
cd ui/litellm-dashboard
../../scripts/with_dashboard_node.sh npm ci
../../scripts/with_dashboard_node.sh npm run build
mkdir -p ../../litellm/proxy/_experimental/out
cp -r out/. ../../litellm/proxy/_experimental/out/
find ../../litellm/proxy/_experimental/out -name '*.html' ! -name index.html | while read -r page; do
mkdir -p "${page%.html}"
mv "${page}" "${page%.html}/index.html"
done
- name: Prepare the isolated database and IdP
run: |
umask 077
bash .github/e2e-stack/start-idp.sh
uv run --no-sync python migrations/run.py > "${RUNNER_TEMP}/mcp-oauth-private/migrations.log" 2>&1
- name: Run every required OAuth variant without retries
run: |
umask 077
uv run --no-sync pytest -c tests/e2e/pytest.ini tests/e2e/mcp/test_mcp_oauth_happy_path_e2e.py \
--rootdir=. --reruns 0 --tb=short -o junit_family=xunit1 \
--junitxml="${RUNNER_TEMP}/mcp-oauth-private/results.xml" \
> "${RUNNER_TEMP}/mcp-oauth-private/pytest.log" 2>&1
- name: Report JUnit results and reject skipped or missing cases
if: always()
run: |
uv run --no-sync python .github/e2e-stack/assert_tests_ran.py \
"${RUNNER_TEMP}/mcp-oauth-private/results.xml" tests/e2e/mcp/test_mcp_oauth_happy_path_e2e.py
- name: Remove private login and logs
if: always()
run: |
docker rm -f e2e-keycloak >/dev/null 2>&1 || true
rm -rf "${RUNNER_TEMP}/mcp-oauth-private"

View file

@ -33,7 +33,7 @@ jobs:
strategy:
fail-fast: false
matrix:
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14", "3.15"]
env:
LITELLM_LOCAL_MODEL_COST_MAP: "True"
steps:
@ -76,6 +76,12 @@ jobs:
- name: Import and CLI checks
run: $IN_NETNS .venv/bin/python .github/scripts/run_merge_smoke.py cli
- name: Routing and UUID compatibility
run: >-
$IN_NETNS .venv/bin/python -m pytest
tests/unit/router_strategy/test_complexity_router.py::TestCustomDimensions
tests/unit/test_uuid_helper.py -q
- name: Proxy startup check
run: $IN_NETNS .venv/bin/python .github/scripts/run_merge_smoke.py proxy-startup --diagnostics-dir "${RUNNER_TEMP}/smoke-diagnostics"

106
.github/workflows/test-rust-db.yml vendored Normal file
View file

@ -0,0 +1,106 @@
name: LiteLLM Rust DB
on:
push:
paths:
- "litellm-rust/crates/db/**"
- "litellm-rust/Cargo.toml"
- "litellm-rust/Cargo.lock"
- "litellm-rust/clippy.toml"
- "litellm-proxy-extras/litellm_proxy_extras/migrations/**"
- "**/schema.prisma"
- ".cargo/**"
- "rust-toolchain.toml"
- "pyproject.toml"
- "uv.lock"
- ".github/actions/setup-uv-with-retries/**"
- ".github/workflows/test-rust-db.yml"
pull_request:
branches:
- main
- "litellm_**"
paths:
- "litellm-rust/crates/db/**"
- "litellm-rust/Cargo.toml"
- "litellm-rust/Cargo.lock"
- "litellm-rust/clippy.toml"
- "litellm-proxy-extras/litellm_proxy_extras/migrations/**"
- "**/schema.prisma"
- ".cargo/**"
- "rust-toolchain.toml"
- "pyproject.toml"
- "uv.lock"
- ".github/actions/setup-uv-with-retries/**"
- ".github/workflows/test-rust-db.yml"
permissions:
contents: read
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: true
env:
CARGO_TERM_COLOR: always
jobs:
db-queries:
runs-on: ubuntu-latest
timeout-minutes: 30
services:
postgres:
image: postgres:16.6@sha256:557fea37a744d5f4c8faab304b0a90858b53ab119735a88c131fd19dab802f36
env:
POSTGRES_DB: litellm
POSTGRES_PASSWORD: postgres
POSTGRES_USER: postgres
ports:
- 5432:5432
options: >-
--health-cmd "pg_isready -U postgres -d litellm"
--health-interval 10s
--health-timeout 5s
--health-retries 5
env:
DATABASE_URL: postgresql://postgres:postgres@localhost:5432/litellm
SQLX_OFFLINE: "false"
defaults:
run:
working-directory: litellm-rust
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
with:
persist-credentials: false
- uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0
with:
python-version: "3.12"
- uses: ./.github/actions/setup-uv-with-retries
with:
version: "0.10.9"
- run: rustup toolchain install --no-self-update
- uses: Swatinem/rust-cache@6323deb102c322ba6fcbdcafc7e3dddab59af2b6 # v2.9.2
with:
workspaces: litellm-rust
cache-on-failure: true
- name: Install Python dependencies
working-directory: .
run: uv sync --frozen --extra extra-proxy
- name: Deploy Prisma migrations
working-directory: .
run: uv run --no-sync prisma migrate deploy --schema litellm-proxy-extras/litellm_proxy_extras/schema.prisma
- name: Install SQLx CLI
run: cargo install sqlx-cli --version 0.9.0 --locked --no-default-features -F postgres
- name: Check SQLx query metadata
working-directory: litellm-rust/crates/db
run: cargo sqlx prepare --check
- name: Test database queries
run: cargo test -p litellm-db --locked -- --include-ignored

View file

@ -16,6 +16,10 @@ on:
- "litellm/proxy/guardrails/**"
- "litellm/utils.py"
- "litellm/ocr/**"
- "litellm/chat_completions/dispatch.py"
- "litellm/embeddings/dispatch.py"
- "litellm/messages/dispatch.py"
- "litellm/responses/dispatch.py"
- "litellm/llms/base_llm/ocr/**"
- "litellm/llms/custom_httpx/llm_http_handler.py"
- "Makefile"
@ -46,6 +50,10 @@ on:
- "litellm/proxy/guardrails/**"
- "litellm/utils.py"
- "litellm/ocr/**"
- "litellm/chat_completions/dispatch.py"
- "litellm/embeddings/dispatch.py"
- "litellm/messages/dispatch.py"
- "litellm/responses/dispatch.py"
- "litellm/llms/base_llm/ocr/**"
- "litellm/llms/custom_httpx/llm_http_handler.py"
- "Makefile"

View file

@ -139,6 +139,7 @@ jobs:
test-path: |-
tests/unit/batch_completion
tests/unit/decisions
tests/unit/extraction
tests/unit/litellm_core_utils
tests/unit/systemone
python-version: "3.12"
@ -215,10 +216,12 @@ jobs:
- shard: llms-providers
test-path: |-
tests/unit/llms/test_api_base_trailing_slash_routes.py
tests/unit/llms/test_cache_control_and_reasoning.py
tests/unit/llms/test_custom_llm.py
tests/unit/llms/test_file_content_block.py
tests/unit/llms/test_file_search_responses.py
tests/unit/llms/test_get_complete_url_trailing_slash.py
tests/unit/llms/test_lifecycle_fix.py
tests/unit/llms/test_oss_decision.py
tests/unit/llms/test_polling_url_origin_match.py
@ -266,6 +269,7 @@ jobs:
tests/unit/llms/featherless_ai
tests/unit/llms/firecrawl
tests/unit/llms/fireworks_ai
tests/unit/llms/friendliai
tests/unit/llms/gdc
tests/unit/llms/gemini
tests/unit/llms/gigachat
@ -321,6 +325,7 @@ jobs:
tests/unit/llms/pass_through
tests/unit/llms/perplexity
tests/unit/llms/pg_vector
tests/unit/llms/pi
tests/unit/llms/publicai
tests/unit/llms/ragflow
tests/unit/llms/recraft
@ -346,6 +351,7 @@ jobs:
tests/unit/llms/tinyfish
tests/unit/llms/together_ai
tests/unit/llms/tool_loop
tests/unit/llms/topaz
tests/unit/llms/triton
tests/unit/llms/v0
tests/unit/llms/valkey
@ -409,6 +415,7 @@ jobs:
tests/unit/embeddings
tests/unit/endpoints
tests/unit/files
tests/unit/fine_tuning
tests/unit/images
tests/unit/interactions
tests/unit/messages
@ -417,6 +424,8 @@ jobs:
tests/unit/rag
tests/unit/realtime_api
tests/unit/rerank_api
tests/unit/search
tests/unit/text_completions
tests/unit/vector_stores
tests/unit/videos
python-version: "3.12"
@ -428,6 +437,13 @@ jobs:
- shard: rust-bridge-harness
test-path: >-
tests/test_litellm_rust/test_dispatch.py
tests/test_litellm_rust/test_public_entries.py
tests/test_litellm_rust/chat_completions/test_dispatch.py
tests/test_litellm_rust/embeddings/test_dispatch.py
tests/test_litellm_rust/messages/test_dispatch.py
tests/test_litellm_rust/ocr/test_dispatch.py
tests/test_litellm_rust/responses/test_dispatch.py
tests/unit/compression
tests/unit/harness
tests/unit/rust_bridge
@ -454,11 +470,11 @@ jobs:
tests/unit/telemetry tests/unit/proxy/telemetry
tests/unit/tracing
python-version: "3.12"
workers: 2
workers: 4
reruns: 0
timeout-minutes: 20
job-timeout-minutes: 60
dist: loadscope
dist: load
- shard: proxy-auth
test-path: |-
@ -563,17 +579,15 @@ jobs:
- shard: mcp-elicitation
test-path: |-
--override-ini=pythonpath=tests
tests/unit/experimental_mcp_client/test_mcp_client.py
tests/unit/proxy/_experimental/mcp_server/test_capabilities.py
tests/unit/proxy/_experimental/mcp_server/test_interactions.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_elicitation_handler.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_server_manager.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_server_tool_calls_and_headers.py
tests/unit/proxy/_experimental/mcp_server/test_operations.py
tests/integration/mcp/test_interactions.py
tests/unit/mcp_client/test_mcp_client.py
tests/unit/proxy/mcp_server/test_capabilities.py
tests/unit/proxy/mcp_server/test_interactions.py
tests/unit/proxy/mcp_server/test_mcp_elicitation_handler.py
tests/unit/proxy/mcp_server/test_mcp_server_manager.py
tests/unit/proxy/mcp_server/test_mcp_server_tool_calls_and_headers.py
tests/unit/proxy/mcp_server/test_operations.py
python-version: "3.12"
workers: 2
workers: 4
reruns: 0
timeout-minutes: 20
job-timeout-minutes: 60
@ -589,7 +603,7 @@ jobs:
tests/unit/proxy/spend_tracking
--ignore=tests/unit/proxy/spend_tracking/test_search_api_logging.py
tests/unit/proxy/pass_through_endpoints
tests/unit/proxy/experimental
tests/unit/proxy/mcp_server/test_tool_registry.py
tests/unit/proxy/common_utils
--ignore=tests/unit/proxy/common_utils/test_cache_aware_routing.py
--ignore=tests/unit/proxy/common_utils/test_check_batch_cost.py
@ -658,6 +672,7 @@ jobs:
- shard: proxy-extras
test-path: >-
tests/unit/litellm_proxy_extras
tests/unit/db_scripts
python-version: "3.12"
workers: 2
reruns: 0
@ -696,7 +711,6 @@ jobs:
tests/unit/responses/litellm_completion_transformation
tests/unit/responses/test_additional_tools.py
tests/unit/responses/test_custom_tool_call.py
tests/unit/responses/test_dispatch.py
tests/unit/responses/test_metadata_codex_callback.py
tests/unit/responses/test_no_duplicate_spend_logs.py
tests/unit/responses/test_null_test_fix.py
@ -797,14 +811,14 @@ jobs:
- shard: mcp-oauth
test-path: |-
tests/unit/proxy/_experimental/mcp_server/test_discoverable_endpoints.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_server_manager.py
tests/unit/proxy/_experimental/mcp_server/test_db_credentials.py
tests/unit/proxy/_experimental/mcp_server/test_gateway_dcr_flow.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_oauth_passthrough_tools.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_max_concurrent_requests.py
tests/unit/proxy/_experimental/mcp_server/faults/test_list_outcomes.py
tests/unit/proxy/_experimental/mcp_server/outbound_credentials
tests/unit/proxy/mcp_server/test_discoverable_endpoints.py
tests/unit/proxy/mcp_server/test_mcp_server_manager.py
tests/unit/proxy/mcp_server/test_db_credentials.py
tests/unit/proxy/mcp_server/test_gateway_dcr_flow.py
tests/unit/proxy/mcp_server/test_mcp_oauth_passthrough_tools.py
tests/unit/proxy/mcp_server/test_mcp_max_concurrent_requests.py
tests/unit/proxy/mcp_server/faults/test_list_outcomes.py
tests/unit/proxy/mcp_server/outbound_credentials
python-version: "3.12"
workers: 4
reruns: 2
@ -814,19 +828,21 @@ jobs:
- shard: mcp-server
test-path: |-
tests/unit/mcp_client
tests/unit/experimental_mcp_client
tests/unit/proxy/mcp_server
tests/unit/proxy/_experimental/mcp_server
tests/unit/responses/mcp
--ignore=tests/unit/experimental_mcp_client/test_mcp_client.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_capabilities.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_interactions.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_mcp_elicitation_handler.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_mcp_server_manager.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_mcp_server_tool_calls_and_headers.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_operations.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_discoverable_endpoints.py
--ignore=tests/unit/proxy/_experimental/mcp_server/test_db_credentials.py
--ignore=tests/unit/proxy/_experimental/mcp_server/outbound_credentials
--ignore=tests/unit/mcp_client/test_mcp_client.py
--ignore=tests/unit/proxy/mcp_server/test_capabilities.py
--ignore=tests/unit/proxy/mcp_server/test_interactions.py
--ignore=tests/unit/proxy/mcp_server/test_mcp_elicitation_handler.py
--ignore=tests/unit/proxy/mcp_server/test_mcp_server_manager.py
--ignore=tests/unit/proxy/mcp_server/test_mcp_server_tool_calls_and_headers.py
--ignore=tests/unit/proxy/mcp_server/test_operations.py
--ignore=tests/unit/proxy/mcp_server/test_discoverable_endpoints.py
--ignore=tests/unit/proxy/mcp_server/test_db_credentials.py
--ignore=tests/unit/proxy/mcp_server/outbound_credentials
python-version: "3.12"
workers: 4
reruns: 0
@ -936,7 +952,7 @@ jobs:
dist: loadscope
steps:
- uses: actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955 # v4.3.0
timeout-minutes: 3
timeout-minutes: ${{ matrix.shard == 'db-and-spend' && 10 || 3 }}
with:
persist-credentials: false
fetch-depth: ${{ matrix.shard == 'db-and-spend' && '0' || '1' }}
@ -1451,7 +1467,6 @@ jobs:
tests/unit/caching/test_redis_connection_pool.py \
tests/unit/caching/test_redis_cluster_cache.py \
tests/unit/caching/test_evicted_client_closer.py \
tests/integration/sdk/test_redis_cluster_iam_auth.py \
--tb=short -vv -rs -p no:rerunfailures -p no:pytest-retry --timeout=120 --durations=20 \
--cov=./litellm --cov-report=xml:coverage.xml

2
.gitignore vendored
View file

@ -105,7 +105,7 @@ litellm_config.yaml
.cursor
litellm/proxy/to_delete_loadtest_work/*
update_model_cost_map.py
tests/unit/proxy/_experimental/mcp_server/test_mcp_server_manager.py
tests/unit/proxy/mcp_server/test_mcp_server_manager.py
scripts/test_vertex_ai_search.py
LAZY_LOADING_IMPROVEMENTS.md
STABILIZATION_TODO.md

View file

@ -29,7 +29,7 @@ A test must only fail when litellm code changes. Never pin facts we don't own (a
`tests/unit/` mirrors `litellm/` in a parallel path (see `tests/unit/AGENTS.md`). Name tests `test_<filename>.py`, but always match the existing test file in the directory you touch — many provider dirs use longer descriptive names (e.g. `test_anthropic_chat_transformation.py`) to avoid ambiguity across sibling folders. For bug fixes, extend the existing mapped test file rather than creating a new one. Only create a new test file for a new feature (provider, endpoint, or transformation module) that has no mapped test yet, following that directory's naming convention (or `test_<filename>.py` if you're the first test there). One focused regression test beats many shallow ones
End-to-end tests belong in `tests/e2e/` and must follow the harness conventions documented in that directory's `AGENTS.md`
End-to-end tests belong in `tests/e2e/`. A rule about one directory's files, like its harness or which CI runs it, goes in that directory's `AGENTS.md`, never only in a README or here. Read a directory's `AGENTS.md` before editing files in it
When creating PRs, target the repository's current default branch for both internal and external / OSS contributions. Check it with `python3 scripts/default_branch.py --branch` instead of assuming a branch name or relying on cached `origin/HEAD`
@ -72,7 +72,7 @@ When referencing or running models (coding, QA'ing, writing docs, writing tests,
Always pull before starting any work. The checkout or worktree may be sitting on a stale branch
If you're an internal contributor, when creating a new PR, the typical flow is to branch off the repository's current default branch and create a branch prefixed with litellm_. Do not create a branch prefixed with claude/ and generally do not have / in your branch names
Internal contributors branch a new PR off the repository's current default branch, with a branch name prefixed with litellm_
Do not add `Co-Authored-By: Claude` or any Claude attribution to commit messages. Never use a `claude/` prefix or put a `/` in a branch name. Do not add "Generated with Claude Code" (or any similar attribution) to PR descriptions or comments. Do not create a new PR/branch off the existing PR to fix/add something that is related and could've just been committed directly to the existing PR's branch

View file

@ -239,6 +239,8 @@ graph LR
7. `DBSpendUpdateWriter.update_database()` queues spend increments to Redis
8. Background job `update_spend` flushes queued spend to PostgreSQL every 60s
Cost lookup falls back to the response model when the provider-reported name cannot be priced. Explicitly registered zero rates remain zero, and custom deployment pricing or an explicit `base_model` retain precedence
### Data Access Layer (Models & Repositories)
Database entities and the operations on them live in two packages at the root of `litellm/` so both the gateway (`proxy/`) and the SDK can use them without importing proxy internals:
@ -332,13 +334,13 @@ Each translation is isolated in its own file, making it easy to test and modify
|--------------|----------|------------------|
| `/v1/chat/completions` | Anthropic | `llms/anthropic/chat/transformation.py` |
| `/v1/chat/completions` | Bedrock Converse | `llms/bedrock/chat/converse_transformation.py` |
| `/v1/chat/completions` | Bedrock Invoke | `llms/bedrock/chat/invoke_transformations/anthropic_claude3_transformation.py` |
| `/v1/chat/completions` | Bedrock Invoke | `llms/bedrock/chat/invoke_transformations/anthropic_claude_transformation.py` |
| `/v1/chat/completions` | Gemini | `llms/gemini/chat/transformation.py` |
| `/v1/chat/completions` | Vertex AI | `llms/vertex_ai/gemini/transformation.py` |
| `/v1/chat/completions` | OpenAI | `llms/openai/chat/gpt_transformation.py` |
| `/v1/messages` (passthrough) | Anthropic | `llms/anthropic/pass_through/messages/transformation.py` |
| `/v1/messages` (passthrough) | Bedrock | `llms/bedrock/messages/invoke_transformations/anthropic_claude3_transformation.py` |
| `/v1/messages` (passthrough) | Vertex AI | `llms/vertex_ai/vertex_ai_partner_models/anthropic/experimental_pass_through/transformation.py` |
| `/v1/messages` (passthrough) | Anthropic | `llms/anthropic/messages/native/transformation.py` |
| `/v1/messages` (passthrough) | Bedrock | `llms/bedrock/messages/invoke_transformations/anthropic_claude_transformation.py` |
| `/v1/messages` (passthrough) | Vertex AI | `llms/vertex_ai/vertex_ai_partner_models/anthropic/messages/transformation.py` |
| Passthrough endpoints | All | `proxy/pass_through_endpoints/llm_provider_handlers/` |
### Example: Debugging prompt caching
@ -346,7 +348,7 @@ Each translation is isolated in its own file, making it easy to test and modify
If `/v1/messages` → Bedrock Converse prompt caching isn't working but Bedrock Invoke works:
1. **Bedrock Converse translation**: `llms/bedrock/chat/converse_transformation.py`
2. **Bedrock Invoke translation**: `llms/bedrock/chat/invoke_transformations/anthropic_claude3_transformation.py`
2. **Bedrock Invoke translation**: `llms/bedrock/chat/invoke_transformations/anthropic_claude_transformation.py`
3. Compare how each handles `cache_control` in `transform_request()`
### How translations work

View file

@ -39,12 +39,12 @@ git clone https://github.com/YOUR_USERNAME/litellm.git
cd litellm
# Create a new branch for your feature (see "Commit and Branch Conventions" below)
git checkout -b feature/your-feature
git checkout -b litellm_your_feature
# Install development dependencies
make install-dev
# Install git hooks that enforce commit + branch conventions (one-time, opt-in)
# Install git hooks that enforce commit conventions (one-time, opt-in)
make install-hooks
# Verify your setup works
@ -55,7 +55,7 @@ That's it! Your local development environment is ready.
## Commit and Branch Conventions
Commits follow [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/) and branches follow [Conventional Branches](https://conventional-branch.github.io/). Run `make install-hooks` once per clone to enable the local git hooks that enforce these — see the [contributor docs](https://docs.litellm.ai/docs/extras/contributing_code#commit-and-branch-conventions) for the full type list, examples, the protected-branch bypass list, and how to opt out.
Commits follow [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/). Internal contributors name branches with the `litellm_` prefix and no `/`, and fork branches can use any name. Run `make install-hooks` once per clone to enable the local hook that enforces the commit convention. See the [contributor docs](https://docs.litellm.ai/docs/extras/contributing_code#commit-and-branch-conventions) for details
### 2. Development Workflow
@ -78,8 +78,8 @@ uv run pytest tests/unit/<your_test_file>.py -v
git add .
git commit -m "feat(scope): your descriptive commit message"
# Push and create a PR (branch must follow Conventional Branches — see above)
git push origin feature/your-feature
# Push and create a PR
git push origin litellm_your_feature
```
## Adding Testing
@ -94,6 +94,8 @@ Add your tests to the [`tests/unit/` directory](https://github.com/BerriAI/litel
- **Only add mocked tests** - no real LLM API calls in this directory
- For integration tests with real APIs, use the appropriate test directories
Tests that call `litellm.register_model` or change `litellm.model_cost` should use the `local_model_cost_map` fixture from `tests/unit/conftest.py`. It provides a fresh catalog and restores the previous map, so pricing and capability overrides cannot leak into later tests
### File Naming Convention
The `tests/unit/` directory follows the same structure as `litellm/`:

View file

@ -1,10 +1,10 @@
# syntax=docker/dockerfile:1.7
# Base image for building
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
# Runtime image
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Pinned by digest like the other base images; bump explicitly on Node upgrades.
ARG UI_BUILD_IMAGE=node:24.19-alpine3.24@sha256:d32cdf619f63fe0471182d08996dd516c6275bb5fd31ae06e55a570bd9e1ad43
@ -61,6 +61,7 @@ RUN apk add --no-cache \
python-3.13-dev \
rust \
openssl \
libssl3 \
openssl-dev \
nodejs \
npm \
@ -120,8 +121,8 @@ RUN sed -i 's/\r$//' docker/entrypoint.sh && chmod +x docker/entrypoint.sh && \
FROM $LITELLM_BUILD_IMAGE AS liteadmin-builder
COPY --from=uvbin /uv /usr/local/bin/uv
RUN apk add --no-cache python-3.13
ADD --checksum=sha256:2f7ae5cdd9d91731c0990e74a58239dc3e3fd2bf28dab23b55eafcdc47aaf87e \
https://github.com/BerriAI/litellm-admin-agent/archive/ef501e94bc9fbacb9233b922abf71427f030408c.tar.gz /tmp/liteadmin.tar.gz
ADD --checksum=sha256:4fb1847e28e2be30ba76dac94f02c4903291326516a0f4ba772f6ffb8b25d05c \
https://github.com/BerriAI/litellm-admin-agent/archive/5661c805ce479a0cf11b842ea133cf8d22cd7e34.tar.gz /tmp/liteadmin.tar.gz
RUN mkdir /tmp/liteadmin && tar xzf /tmp/liteadmin.tar.gz --strip-components=1 -C /tmp/liteadmin && \
uv venv /opt/liteadmin --python python3.13 && \
uv pip install --python /opt/liteadmin/bin/python --require-hashes -r /tmp/liteadmin/requirements.txt && \
@ -141,7 +142,7 @@ USER root
RUN echo "https://packages.wolfi.dev/os" >> /etc/apk/repositories
# node (without npm) is required by the prisma CLI at runtime
RUN apk add --no-cache bash openssl tzdata nodejs python-3.13 libsndfile libevent
RUN apk add --no-cache bash openssl libssl3 tzdata nodejs python-3.13 libsndfile libevent
COPY --from=pgbouncer-builder /usr/local/bin/pgbouncer /usr/local/bin/pgbouncer
WORKDIR /app

View file

@ -22,7 +22,7 @@ help:
@echo " make install-proxy-dev-ci - Install proxy dev dependencies (CI-compatible)"
@echo " make install-test-deps - Install the full local test environment"
@echo " make install-helm-unittest - Install helm unittest plugin"
@echo " make install-hooks - Install git hooks (Conventional Commits + Branches)"
@echo " make install-hooks - Install git hooks (Conventional Commits)"
@echo " make check - Run CI-equivalent lint on staged files, or on the diff vs the base branch when nothing is staged"
@echo " make pre-commit - Legacy alias for make check"
@echo " make format - Apply ruff format code formatting"
@ -120,7 +120,7 @@ install-helm-unittest:
helm plugin install https://github.com/helm-unittest/helm-unittest --version v0.8.2; \
}
# Install git hooks that enforce Conventional Commits and Conventional Branches.
# Install git hooks that enforce Conventional Commits.
# Opt-in: not chained into install-dev.
install-hooks:
./scripts/install_git_hooks.sh
@ -290,7 +290,7 @@ test-rust-extension:
"$$temporary/venv/bin/python" -I -m mypy.stubtest \
--mypy-config-file tests/unit/rust_bridge/stubtest.ini \
litellm.rust_bridge._native && \
LITELLM_RUST=1 LITELLM_LOCAL_MODEL_COST_MAP=True \
LITELLM_LOCAL_MODEL_COST_MAP=True \
"$$temporary/venv/bin/python" -I -m pytest --import-mode=importlib -m requires_rust_extension tests/test_litellm_rust
rust-sqlx-prepare:
@ -313,7 +313,7 @@ test-unit-proxy-core: install-test-deps
$(UV_RUN) pytest tests/unit/proxy/auth tests/unit/proxy/client tests/unit/proxy/db tests/unit/proxy/hooks tests/unit/proxy/policy_engine --ignore=tests/unit/proxy/db/db_transaction_queue/test_e2e_pod_lock_manager.py --ignore=tests/unit/proxy/db/test_update_daily_tag_spend.py --tb=short -vv -n 4 --durations=20
test-unit-proxy-misc: install-test-deps
$(UV_RUN) pytest tests/unit/proxy/agent_endpoints tests/unit/proxy/anthropic_endpoints tests/unit/proxy/common_utils --ignore=tests/unit/proxy/common_utils/test_cache_aware_routing.py --ignore=tests/unit/proxy/common_utils/test_check_batch_cost.py --ignore=tests/unit/proxy/common_utils/test_check_responses_cost.py --ignore=tests/unit/proxy/common_utils/test_proxy_encrypt_decrypt.py --ignore=tests/unit/proxy/common_utils/test_realtime_cache.py tests/unit/proxy/discovery_endpoints tests/unit/proxy/experimental tests/unit/proxy/google_endpoints tests/unit/proxy/health_endpoints tests/unit/proxy/image_endpoints tests/unit/proxy/middleware --ignore=tests/unit/proxy/middleware/test_request_size_limit_middleware.py tests/unit/proxy/openai_files_endpoint tests/unit/proxy/pass_through_endpoints tests/unit/proxy/prompts tests/unit/proxy/public_endpoints tests/unit/proxy/response_api_endpoints tests/unit/proxy/shutdown tests/unit/proxy/spend_tracking --ignore=tests/unit/proxy/spend_tracking/test_search_api_logging.py tests/unit/proxy/ui_crud_endpoints tests/unit/proxy/vector_store_endpoints tests/unit/proxy/_experimental/mcp_server/test_mcp_server_tool_calls_and_headers.py --ignore=tests/unit/proxy/google_endpoints/test_gemini_agents_endpoints.py --ignore=tests/unit/proxy/google_endpoints/test_google_endpoint_routing.py --ignore=tests/unit/proxy/google_endpoints/test_google_gemini_proxy_request.py --ignore=tests/unit/proxy/public_endpoints/test_blog_posts_endpoint.py --tb=short -vv -n 4 --durations=20
$(UV_RUN) pytest tests/unit/proxy/agent_endpoints tests/unit/proxy/anthropic_endpoints tests/unit/proxy/common_utils --ignore=tests/unit/proxy/common_utils/test_cache_aware_routing.py --ignore=tests/unit/proxy/common_utils/test_check_batch_cost.py --ignore=tests/unit/proxy/common_utils/test_check_responses_cost.py --ignore=tests/unit/proxy/common_utils/test_proxy_encrypt_decrypt.py --ignore=tests/unit/proxy/common_utils/test_realtime_cache.py tests/unit/proxy/discovery_endpoints tests/unit/proxy/mcp_server/test_tool_registry.py tests/unit/proxy/google_endpoints tests/unit/proxy/health_endpoints tests/unit/proxy/image_endpoints tests/unit/proxy/middleware --ignore=tests/unit/proxy/middleware/test_request_size_limit_middleware.py tests/unit/proxy/openai_files_endpoint tests/unit/proxy/pass_through_endpoints tests/unit/proxy/prompts tests/unit/proxy/public_endpoints tests/unit/proxy/response_api_endpoints tests/unit/proxy/shutdown tests/unit/proxy/spend_tracking --ignore=tests/unit/proxy/spend_tracking/test_search_api_logging.py tests/unit/proxy/ui_crud_endpoints tests/unit/proxy/vector_store_endpoints tests/unit/proxy/mcp_server/test_mcp_server_tool_calls_and_headers.py --ignore=tests/unit/proxy/google_endpoints/test_gemini_agents_endpoints.py --ignore=tests/unit/proxy/google_endpoints/test_google_endpoint_routing.py --ignore=tests/unit/proxy/google_endpoints/test_google_gemini_proxy_request.py --ignore=tests/unit/proxy/public_endpoints/test_blog_posts_endpoint.py --tb=short -vv -n 4 --durations=20
test-unit-proxy-root: install-test-deps
$(UV_RUN) pytest tests/unit/proxy/test_*.py --ignore=tests/unit/proxy/test_aproxy_startup.py --ignore=tests/unit/proxy/test_credential_slot_registry.py --ignore=tests/unit/proxy/test_custom_callback_input.py --ignore=tests/unit/proxy/test_custom_logger_s3_gcs.py --ignore=tests/unit/proxy/test_custom_tokenizer_bug.py --ignore=tests/unit/proxy/test_db_schema_changes.py --ignore=tests/unit/proxy/test_deprecated_key_grace_period.py --ignore=tests/unit/proxy/test_get_favicon.py --ignore=tests/unit/proxy/test_get_image.py --ignore=tests/unit/proxy/test_prisma_client_backoff_retry.py --ignore=tests/unit/proxy/test_prompt_test_endpoint.py --ignore=tests/unit/proxy/test_proxy_config_unit_test.py --ignore=tests/unit/proxy/test_proxy_custom_auth.py --ignore=tests/unit/proxy/test_proxy_reject_logging.py --ignore=tests/unit/proxy/test_proxy_server.py --ignore=tests/unit/proxy/test_proxy_setting_guardrails.py --ignore=tests/unit/proxy/test_proxy_token_counter.py --ignore=tests/unit/proxy/test_proxy_utils.py --ignore=tests/unit/proxy/test_reducto_ocr_route.py --ignore=tests/unit/proxy/test_response_polling_pre_call_checks.py --ignore=tests/unit/proxy/test_server_root_path.py --ignore=tests/unit/proxy/test_ui_path_detection.py --ignore=tests/unit/proxy/test_unit_test_proxy_hooks.py --ignore=tests/unit/proxy/test_update_spend.py --ignore=tests/unit/proxy/test_zero_cost_model_budget_bypass.py --tb=short -vv -n 4 --durations=20
@ -325,7 +325,7 @@ test-unit-core-utils: install-test-deps
$(UV_RUN) pytest tests/unit/litellm_core_utils --tb=short -vv -n 2 --durations=20
test-unit-other: install-test-deps
$(UV_RUN) pytest tests/unit/caching tests/unit/responses tests/unit/secret_managers tests/unit/vector_stores tests/unit/a2a_protocol tests/unit/completion_extras tests/unit/containers tests/unit/enterprise tests/unit/experimental_mcp_client tests/unit/google_genai tests/unit/images tests/unit/interactions tests/unit/router_strategy tests/unit/router_utils tests/unit/types --tb=short -vv -n 4 --durations=20
$(UV_RUN) pytest tests/unit/caching tests/unit/responses tests/unit/secret_managers tests/unit/vector_stores tests/unit/a2a_protocol tests/unit/completion_extras tests/unit/containers tests/unit/enterprise tests/unit/mcp_client tests/unit/google_genai tests/unit/images tests/unit/interactions tests/unit/router_strategy tests/unit/router_utils tests/unit/types --tb=short -vv -n 4 --durations=20
test-unit-root: install-test-deps
$(UV_RUN) pytest tests/unit/test_*.py --tb=short -vv -n 4 --durations=20

View file

@ -223,7 +223,7 @@ async with httpx.AsyncClient(headers=headers, timeout=60.0) as http_client:
```python
from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
from litellm import experimental_mcp_client
from litellm import mcp_client
import litellm
server_params = StdioServerParameters(command="python", args=["mcp_server.py"])
@ -233,7 +233,7 @@ async with stdio_client(server_params) as (read, write):
await session.initialize()
# Load MCP tools in OpenAI format
tools = await experimental_mcp_client.load_mcp_tools(session=session, format="openai")
tools = await mcp_client.load_mcp_tools(session=session, format="openai")
# Use with any LiteLLM model
response = await litellm.acompletion(
@ -287,7 +287,7 @@ For MCP OAuth, an upstream may advertise dynamic client registration but refuse
</details>
<details>
<summary><b>Agents</b> - Run Claude Code, Codex, OpenCode or Deep Agents on any model (Python SDK)</summary>
<summary><b>Agents</b> - Run Claude Code, Codex, OpenCode, pi or Deep Agents on any model (Python SDK)</summary>
### Python SDK - Agents
@ -296,7 +296,7 @@ import litellm
from litellm import Harness, sandbox
result = litellm.agent(
Harness.CLAUDE_CODE, # or Harness.CODEX, Harness.OPENCODE, Harness.DEEPAGENTS
Harness.CLAUDE_CODE, # or Harness.CODEX, Harness.OPENCODE, Harness.PI, Harness.DEEPAGENTS
"Find why tests/test_router.py is flaky and fix it.",
sandbox=sandbox.local("./repo"),
model="litellm_proxy/claude-sonnet-4-5", # a model group on your AI Gateway
@ -305,7 +305,7 @@ result = litellm.agent(
print(result.text, result.cost, [f.path for f in result.files])
```
Set `LITELLM_PROXY_API_BASE` and `LITELLM_PROXY_API_KEY` and every model call the agent makes goes through your AI Gateway, tagged `harness,claude_code`. Drop the `litellm_proxy/` prefix to call a provider directly. Install `starlette uvicorn` plus the agent's CLI (`claude`, `codex` or `opencode`), or `deepagents langchain-litellm` for Deep Agents.
Set `LITELLM_PROXY_API_BASE` and `LITELLM_PROXY_API_KEY` and every model call the agent makes goes through your AI Gateway, tagged `harness,claude_code`. Drop the `litellm_proxy/` prefix to call a provider directly. Install `starlette uvicorn` plus the agent's CLI (`claude`, `codex`, `opencode` or `pi`), or `deepagents langchain-litellm` for Deep Agents.
[**Docs: Agent Harnesses**](https://docs.litellm.ai/docs/harness)
@ -384,6 +384,7 @@ Set `LITELLM_PROXY_API_BASE` and `LITELLM_PROXY_API_KEY` and every model call th
| [ModelScope (`modelscope`)](https://docs.litellm.ai/docs/providers/modelscope) | ✅ | ✅ | ✅ | | ✅ | | | | | |
| [Moonshot (`moonshot`)](https://docs.litellm.ai/docs/providers/moonshot) | ✅ | ✅ | ✅ | | | | | | | |
| [Morph (`morph`)](https://docs.litellm.ai/docs/providers/morph) | ✅ | ✅ | ✅ | | | | | | | |
| [Nace (`nace`)](https://console.nace.ai/docs/drex) | | | | | | | | | | |
| [Nebius AI Studio (`nebius`)](https://docs.litellm.ai/docs/providers/nebius) | ✅ | ✅ | ✅ | ✅ | | | | | | |
| [NLP Cloud (`nlp_cloud`)](https://docs.litellm.ai/docs/providers/nlp_cloud) | ✅ | ✅ | ✅ | | | | | | | |
| [Novita AI (`novita`)](https://novita.ai/models/llm?utm_source=github_litellm&utm_medium=github_readme&utm_campaign=github_link) | ✅ | ✅ | ✅ | | | | | | | |
@ -615,15 +616,15 @@ The Terraform modules live at [`terraform/litellm/aws/`](./terraform/litellm/aws
### Verify Docker Image Signatures
All LiteLLM Docker images published to GHCR are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0).
All LiteLLM Docker images published to GHCR are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Releases are signed with the key introduced in [commit `e82a3c6`](https://github.com/BerriAI/litellm/commit/e82a3c646908ba0a6ee0beddc4310d7a6cac4065).
**Verify using the pinned commit hash (recommended):**
A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key:
A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the published signing key:
```bash
cosign verify \
--key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \
--key https://raw.githubusercontent.com/BerriAI/litellm/e82a3c646908ba0a6ee0beddc4310d7a6cac4065/cosign.pub \
ghcr.io/berriai/litellm:<release-tag>
```

View file

@ -79,7 +79,7 @@ To qualify for a bounty, reports must include clear reproduction steps, a reprod
- Audit Logs with retention policy
- Control Allowed IP Addresses that can access your Cloud LiteLLM Instance
For security inquiries, please contact us at support@berri.ai
For security inquiries, please contact us at security@berri.ai
#### Supported data regions for LiteLLM Cloud

View file

@ -1,9 +1,36 @@
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
FROM $UV_IMAGE AS uvbin
# ---------- Rust extension ----------
# Copies only the native build's inputs, so the fat-LTO compile stays cached across Python-only changes
FROM $LITELLM_BUILD_IMAGE AS rust-native
WORKDIR /app
USER root
COPY --from=uvbin /uv /uvx /usr/local/bin/
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
ENV UV_PYTHON_DOWNLOADS=0
COPY pyproject.toml README.md rust-toolchain.toml ./
COPY .cargo ./.cargo
COPY litellm-rust ./litellm-rust
RUN --mount=type=cache,target=/root/.cache/uv \
mkdir -p litellm/rust_bridge /wheels && \
touch litellm/__init__.py litellm/rust_bridge/__init__.py && \
uvx --python python3.13 --from maturin==1.15.0 --with puccinialin==0.1.14 \
python -c "import maturin; maturin.build_wheel('/wheels')"
# ---------- Builder ----------
FROM $LITELLM_BUILD_IMAGE AS builder
@ -16,7 +43,7 @@ COPY --from=uvbin /uv /uvx /usr/local/bin/
# instead of nodeenv downloading one whose dynamic deps may not be in Wolfi
# (e.g. Node 26.2.0 needs libatomic). Retry for transient apk.cgr.dev flakes.
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev libsndfile nodejs npm && break; \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl libssl3 openssl-dev libsndfile nodejs npm && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -54,7 +81,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
COPY . .
RUN --mount=type=cache,target=/root/.cache/uv \
uv sync --frozen --no-default-groups --no-editable \
uv sync --frozen --no-install-project --no-default-groups --no-editable \
--extra proxy \
--extra proxy-runtime \
--group admin-mcp \
@ -63,6 +90,9 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra saml \
--python python3.13
RUN --mount=type=bind,from=rust-native,source=/wheels,target=/wheels \
uv pip install --python /app/.venv/bin/python --no-deps /wheels/litellm-*.whl
RUN cp "$(python -c 'import sysconfig; print(sysconfig.get_paths()["purelib"])')"/litellm/rust_bridge/_native*.so litellm/rust_bridge/
RUN HOME=/opt/prisma XDG_CACHE_HOME=/opt/prisma/.cache PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries \
@ -79,7 +109,7 @@ ENV LITELLM_RELEASE_TAG=${LITELLM_RELEASE_TAG}
USER root
RUN for i in 1 2 3; do \
apk add --no-cache bash openssl tzdata python-3.13 libsndfile libatomic && break; \
apk add --no-cache bash openssl libssl3 tzdata python-3.13 libsndfile libatomic && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -94,7 +124,12 @@ ENV HOME=/home/nonroot \
PYTHONUNBUFFERED=1 \
PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries
COPY --from=builder --chown=nonroot:nonroot /app /app
COPY --from=builder --chown=nonroot:nonroot /app/.venv /app/.venv
COPY --from=builder --chown=nonroot:nonroot /app/litellm /app/litellm
COPY --from=builder --chown=nonroot:nonroot /app/enterprise /app/enterprise
COPY --from=builder --chown=nonroot:nonroot /app/backend /app/backend
COPY --from=builder --chown=nonroot:nonroot /app/docker/component_entrypoint.sh /app/docker/component_entrypoint.sh
COPY --from=builder --chown=nonroot:nonroot /app/schema.prisma /app/schema.prisma
COPY --from=builder /opt/prisma /opt/prisma
RUN find /app/.venv -type f -path "*/tornado/test/*" -delete && \

View file

@ -165,6 +165,7 @@ BACKEND_EXACT_PATHS: frozenset[str] = frozenset(
BACKEND_MOUNT_PATHS: frozenset[str] = frozenset(
{
"/admin",
"/metrics",
"/swagger", # API documentation static assets belong to the backend
"/mcp", # lazily-mounted MCP sub-app serves on the backend component
}

View file

@ -1,11 +1,10 @@
"""Guard the cost map on pull requests.
Every pull request whose diff against its merge base touches one of the three cost map files gets the file
checks: the files parse, the backup copy matches the root file, and the JSON schema is in sync and validates the
map. A pull request that leaves all three untouched skips them, since merging it keeps the base branch's copies
and its head tree only carries whatever state the branch was cut from. Pull requests from the cost map sync bot
(branches named litellm_cost_map_sync_*) always get the file checks and additionally may only touch those three
files and may only add or update models.
Every pull request whose diff against its merge base touches one of the two cost map files gets the file
checks: the files parse and the backup copy matches the root file. A pull request that leaves both untouched
skips them, since merging it keeps the base branch's copies and its head tree only carries whatever state the
branch was cut from. Pull requests from the cost map sync bot (branches named litellm_cost_map_sync_*) always
get the file checks and additionally may only touch those two files and may only add or update models.
"""
from __future__ import annotations
@ -19,13 +18,11 @@ from dataclasses import dataclass
from types import MappingProxyType
from typing import Final
from generate_model_prices_schema import SPECIAL_ROOT_KEYS, build_schema, render, validation_errors
COST_MAP_PATH: Final = "model_prices_and_context_window.json"
BACKUP_PATH: Final = "litellm/model_prices_and_context_window_backup.json"
SCHEMA_PATH: Final = "model_prices_and_context_window.schema.json"
GUARDED_PATHS: Final = (COST_MAP_PATH, BACKUP_PATH, SCHEMA_PATH)
GUARDED_PATHS: Final = (COST_MAP_PATH, BACKUP_PATH)
BOT_BRANCH_PREFIX: Final = "litellm_cost_map_sync_"
SPECIAL_ROOT_KEYS: Final = frozenset({"sample_spec", "fallback_generalizations"})
EVALUATION_OUTPUT_PRICE_SOURCES: Final[Mapping[str, str]] = MappingProxyType({})
CostMap = dict[str, object]
@ -35,7 +32,6 @@ CostMap = dict[str, object]
class Snapshot:
cost_map: str
backup: str
schema: str
def _parse_object(text: str, path: str) -> CostMap | str:
@ -46,13 +42,6 @@ def _parse_object(text: str, path: str) -> CostMap | str:
return parsed if isinstance(parsed, dict) else f"{path} must be a JSON object at the root"
def _rendered_schema(cost_map: CostMap) -> str:
try:
return render(build_schema(cost_map))
except SystemExit as error:
return str(error)
def _evaluation_output_price_failure(
key: str, entry: object, evaluation_output_price_sources: Mapping[str, str]
) -> str | None:
@ -78,29 +67,13 @@ def _file_failures(
*,
evaluation_output_price_sources: Mapping[str, str] = EVALUATION_OUTPUT_PRICE_SOURCES,
) -> tuple[str, ...]:
schema_text: Final = _rendered_schema(head_map)
if not schema_text.startswith("{"):
return (schema_text,)
backup_failure: Final = (
()
if head.backup == head.cost_map
else (f"{BACKUP_PATH} differs from {COST_MAP_PATH}; copy the root file over it",)
)
schema_failure: Final = (
()
if head.schema == schema_text
else (
f"{SCHEMA_PATH} is out of sync with {COST_MAP_PATH}; "
"run `python ci_cd/generate_model_prices_schema.py` and commit the result",
)
)
return (
*backup_failure,
*schema_failure,
*(
f"{COST_MAP_PATH} does not validate against its schema: {error}"
for error in validation_errors(head_map, json.loads(schema_text))[:20]
),
*(
failure
for key, entry in head_map.items()

View file

@ -1,409 +0,0 @@
from __future__ import annotations
import json
import sys
from pathlib import Path
from typing import Final, Optional
import jsonschema
REPO_ROOT = Path(__file__).parent.parent
PRICES_PATH = REPO_ROOT / "model_prices_and_context_window.json"
SCHEMA_PATH = REPO_ROOT / "model_prices_and_context_window.schema.json"
SPECIAL_ROOT_KEYS = frozenset({"sample_spec", "fallback_generalizations"})
JsonSchema = dict
NONNEG_NUMBER: JsonSchema = {"type": "number", "minimum": 0}
NONNEG_INTEGER: JsonSchema = {"type": "integer", "minimum": 0}
BOOLEAN: JsonSchema = {"type": "boolean"}
STRING: JsonSchema = {"type": "string"}
TIME_WINDOW: Final[JsonSchema] = {"type": "string", "pattern": r"^([01]\d|2[0-3]):[0-5]\d-([01]\d|2[0-3]):[0-5]\d$"}
WEEKDAY_PATTERN: Final = (
r"(?i)^(mon|monday|tue|tues|tuesday|wed|wednesday|thu|thur|thurs|thursday|fri|friday|sat|saturday|sun|sunday)$"
)
EXTRA_BOOLEAN_KEYS = frozenset(
{
"gemini_native_audio",
"gemini_audio_only_live",
"uses_embed_content",
"use_openai_responses_path",
"bedrock_converse_supports_strict_tools",
"thinking_always_on",
}
)
HOURS_UTC: Final[JsonSchema] = {
"description": 'UTC "HH:MM-HH:MM" window, or a list of them; a window may wrap past midnight.',
"oneOf": [TIME_WINDOW, {"type": "array", "items": TIME_WINDOW, "minItems": 1}],
}
OFF_PEAK_WINDOW: Final[JsonSchema] = {
"type": "object",
"properties": {
"hours_utc": HOURS_UTC,
"weekdays": {
"type": "array",
"description": "ISO-8601 weekday numbers (1 = Monday .. 7 = Sunday) or English day names the window applies on.",
"items": {
"oneOf": [
{"type": "integer", "minimum": 1, "maximum": 7},
{"type": "string", "pattern": WEEKDAY_PATTERN},
]
},
"minItems": 1,
},
},
"required": ["hours_utc"],
"additionalProperties": False,
}
OBJECT_KEYS: dict[str, JsonSchema] = {
"off_peak_pricing": {
"type": "object",
"description": "Rates that replace the same-named base fields while the request falls inside the stated UTC windows.",
"properties": {
"hours_utc": HOURS_UTC,
"windows": {"type": "array", "items": OFF_PEAK_WINDOW, "minItems": 1},
"weekday_timezone": {
"type": "string",
"description": "IANA zone the weekdays of each window are read on; defaults to UTC.",
},
"input_cost_per_token": NONNEG_NUMBER,
"output_cost_per_token": NONNEG_NUMBER,
"output_cost_per_reasoning_token": NONNEG_NUMBER,
"cache_read_input_token_cost": NONNEG_NUMBER,
"cache_creation_input_token_cost": NONNEG_NUMBER,
},
"anyOf": [{"required": ["hours_utc"]}, {"required": ["windows"]}],
"additionalProperties": False,
},
"search_context_cost_per_query": {
"type": "object",
"description": "USD cost per web search query, keyed by search context size.",
"properties": {
"search_context_size_low": NONNEG_NUMBER,
"search_context_size_medium": NONNEG_NUMBER,
"search_context_size_high": NONNEG_NUMBER,
},
"additionalProperties": False,
},
"guardrail_cost_per_unit": {
"type": "object",
"description": "USD cost per billable guardrail unit, keyed by the provider's usage counter name (e.g. Bedrock's contentPolicyUnits).",
"additionalProperties": NONNEG_NUMBER,
},
"metadata": {
"type": "object",
"description": "Free-form notes about the entry (e.g. pricing derivation).",
},
"provider_specific_entry": {
"type": "object",
"description": "Provider-internal routing hints (e.g. bedrock_invocation_schema).",
},
}
ARRAY_KEYS: dict[str, JsonSchema] = {
"supported_audio_formats": {
"type": "array",
"description": "Audio container formats the model can return.",
"items": {"type": "string", "enum": ["mp3", "wav"]},
},
"supported_endpoints": {
"type": "array",
"description": "OpenAI-style API routes this model can be called through, e.g. /v1/chat/completions.",
"items": STRING,
},
"supported_modalities": {
"type": "array",
"description": "Input modalities the model accepts.",
"items": {"type": "string", "enum": ["text", "image", "audio", "video"]},
},
"supported_output_modalities": {
"type": "array",
"description": "Output modalities the model can produce.",
"items": {"type": "string", "enum": ["text", "image", "audio", "video", "code"]},
},
"reasoning_effort_levels": {
"type": "array",
"description": "Exact reasoning_effort levels this deployment accepts; wins over supports_* flags.",
"items": {"type": "string", "enum": ["none", "minimal", "low", "medium", "high", "xhigh", "max"]},
},
"supported_regions": {
"type": "array",
"description": "Cloud regions the model is available in ('global' or region ids).",
"items": STRING,
},
"tiered_pricing": {
"type": "array",
"description": "Context-length or result-count tiered rates; each tier's costs apply within its range.",
"items": {
"type": "object",
"properties": {
"range": {
"type": "array",
"description": "[min, max] prompt-token span this tier applies to.",
"items": NONNEG_NUMBER,
"minItems": 2,
"maxItems": 2,
},
"max_results_range": {
"type": "array",
"description": "[min, max] result-count span this tier applies to (search models).",
"items": NONNEG_NUMBER,
"minItems": 2,
"maxItems": 2,
},
"input_cost_per_token": NONNEG_NUMBER,
"output_cost_per_token": NONNEG_NUMBER,
"output_cost_per_reasoning_token": NONNEG_NUMBER,
"cache_read_input_token_cost": NONNEG_NUMBER,
"cache_creation_input_token_cost": NONNEG_NUMBER,
"input_cost_per_query": NONNEG_NUMBER,
},
"additionalProperties": False,
},
},
}
INTEGER_KEYS: dict[str, JsonSchema] = {
"max_tokens": {
**NONNEG_INTEGER,
"description": "Legacy field: max output tokens if the provider specifies it, else max input tokens.",
},
"max_input_tokens": {
**NONNEG_INTEGER,
"description": "Maximum prompt/context tokens the model accepts.",
},
"max_output_tokens": {
**NONNEG_INTEGER,
"description": "Maximum tokens the model can generate in one response.",
},
"output_vector_size": {
**NONNEG_INTEGER,
"description": "Embedding dimension for embedding models.",
},
"prompt_cache_min_tokens": {
**NONNEG_INTEGER,
"description": "Smallest prefix the provider will actually cache; absent means the provider default applies.",
},
"tpm": {**NONNEG_INTEGER, "description": "Provider default tokens-per-minute limit."},
"rpm": {**NONNEG_INTEGER, "description": "Provider default requests-per-minute limit."},
}
NUMBER_KEYS: dict[str, JsonSchema] = {
"regional_processing_uplift_multiplier_eu": {
"type": "number",
"minimum": 1,
"description": "Multiplier applied to all token costs for EU data residency (e.g. 1.10 = +10%).",
},
"regional_processing_uplift_multiplier_us": {
"type": "number",
"minimum": 1,
"description": "Multiplier applied to all token costs for US data residency (e.g. 1.10 = +10%).",
},
"regional_endpoint_uplift_multiplier": {
"type": "number",
"minimum": 1,
"description": "Multiplier applied to all token costs when served from a non-global Vertex AI endpoint (e.g. 1.10 = +10%).",
},
}
COST_DESCRIPTIONS: dict[str, str] = {
"input_cost_per_token": "USD per prompt token.",
"output_cost_per_token": "USD per generated token.",
"output_cost_per_reasoning_token": "USD per reasoning/thinking token, when billed separately.",
"google_maps_grounding_cost_per_query": (
"USD per Grounding with Google Maps request; billed per query or per prompt per web_search_billing_unit."
),
"cache_creation_input_token_cost": "USD per token written to the provider's prompt cache.",
"cache_read_input_token_cost": "USD per prompt token served from the provider's prompt cache.",
"input_cost_per_token_batches": "USD per prompt token via the provider's batch API.",
"output_cost_per_token_batches": "USD per generated token via the provider's batch API.",
}
def cost_description(key: str) -> Optional[str]:
if key in COST_DESCRIPTIONS:
return COST_DESCRIPTIONS[key]
if key.endswith("_flex"):
return "Flex service-tier rate for the same-named base field."
if key.endswith("_priority"):
return "Priority service-tier rate for the same-named base field."
if "_above_" in key:
return "Rate applied once the prompt exceeds the token threshold in the field name."
return None
def cost_schema(key: str) -> JsonSchema:
description = cost_description(key)
return {**NONNEG_NUMBER, "description": description} if description else dict(NONNEG_NUMBER)
def string_key_schemas(modes: tuple) -> dict[str, JsonSchema]:
return {
"litellm_provider": {
"type": "string",
"description": "LiteLLM provider slug; one of https://docs.litellm.ai/docs/providers.",
},
"mode": {
"type": "string",
"description": "Primary API surface / task type of the model.",
"enum": list(modes),
},
"source": {
"type": "string",
"description": "URL of the provider pricing/model page this entry was taken from.",
},
"deprecation_date": {
"type": "string",
"description": "Date the provider deprecates the model, YYYY-MM-DD.",
"format": "date",
"pattern": "^\\d{4}-(0[1-9]|1[0-2])-(0[1-9]|[12]\\d|3[01])$",
},
"web_search_billing_unit": {
"type": "string",
"description": "Whether web search is billed per query or per prompt.",
"enum": ["per_query", "per_prompt"],
},
"bedrock_output_config_effort_ceiling": {
"type": "string",
"description": "Highest reasoning effort the Bedrock output_config accepts for this model.",
"enum": ["low", "medium", "high", "max", "xhigh"],
},
"default_reasoning_effort": {
"type": "string",
"description": (
"Reasoning effort the provider applies when the request omits reasoning_effort. "
"Gates whether a non-default temperature or the top_p/logprobs sampling params are "
"accepted, which hold only when the effort resolves to 'none'."
),
"enum": ["none", "minimal", "low", "medium", "high", "xhigh"],
},
"comment": STRING,
"audio_transcription_config": STRING,
"vertex_ai_audio_api": {
"type": "string",
"enum": ["lyria_predict", "lyria_interactions"],
},
}
def classify(key: str, modes: tuple) -> Optional[JsonSchema]:
curated = {**OBJECT_KEYS, **ARRAY_KEYS, **string_key_schemas(modes), **INTEGER_KEYS, **NUMBER_KEYS}
if key in curated:
return curated[key]
if key.startswith("supports_") or key in EXTRA_BOOLEAN_KEYS:
return BOOLEAN
if "cost" in key:
return cost_schema(key)
return None
def build_schema(prices: dict) -> JsonSchema:
entries = {name: entry for name, entry in prices.items() if name not in SPECIAL_ROOT_KEYS}
all_keys = tuple(sorted({key for entry in entries.values() for key in entry}))
modes = tuple(sorted({entry["mode"] for entry in entries.values() if "mode" in entry}))
unclassified = tuple(key for key in all_keys if classify(key, modes) is None)
if unclassified:
raise SystemExit(
f"Unclassified keys in {PRICES_PATH.name}: {', '.join(unclassified)}. "
f"Add them to the key tables in {Path(__file__).name} and rerun it."
)
entry_properties = {key: classify(key, modes) for key in all_keys}
return {
"$schema": "https://json-schema.org/draft/2020-12/schema",
"title": "LiteLLM model_prices_and_context_window.json",
"description": (
"Schema for LiteLLM's model price and context window registry "
"(https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json). "
"Every top-level key except 'sample_spec' and 'fallback_generalizations' is a model id, "
"optionally prefixed with its provider (e.g. 'azure/gpt-5.4'), mapping to a model entry. "
"All costs are USD per unit. New optional fields are added regularly, so consumers should "
"ignore unknown fields rather than reject them."
),
"type": "object",
"properties": {
"sample_spec": {
"type": "object",
"description": (
"Documentation placeholder illustrating the entry shape; not a real model and not "
"schema-conformant (several values are prose)."
),
},
"fallback_generalizations": {
"type": "object",
"description": "Regex rules that generalize unknown model ids to known families; not a model entry.",
"properties": {
"rules": {
"type": "array",
"items": {
"type": "object",
"properties": {
"name": STRING,
"pattern": STRING,
"description": STRING,
},
"required": ["name", "pattern"],
"additionalProperties": True,
},
}
},
"additionalProperties": False,
},
},
"additionalProperties": {"$ref": "#/$defs/modelEntry"},
"$defs": {
"modelEntry": {
"type": "object",
"description": (
"Pricing, limits, and capability flags for one model. Fields other than litellm_provider "
"are optional; boolean capability flags are simply omitted when unknown or false."
),
"required": ["litellm_provider"],
"properties": entry_properties,
"additionalProperties": True,
}
},
}
def render(schema: JsonSchema) -> str:
return json.dumps(schema, indent=2) + "\n"
def validation_errors(prices: dict, schema: JsonSchema) -> tuple:
validator = jsonschema.Draft202012Validator(schema, format_checker=jsonschema.Draft202012Validator.FORMAT_CHECKER)
return tuple(
f"{'.'.join(str(part) for part in error.absolute_path)}: {error.message}"
for error in validator.iter_errors(prices)
)
def main() -> int:
check = "--check" in sys.argv[1:]
prices = json.loads(PRICES_PATH.read_text())
rendered = render(build_schema(prices))
errors = validation_errors(prices, json.loads(rendered))
if errors:
print(f"{PRICES_PATH.name} does not validate against the generated schema:")
print("\n".join(errors[:20]))
return 1
if not check:
SCHEMA_PATH.write_text(rendered)
print(f"wrote {SCHEMA_PATH}")
return 0
if not SCHEMA_PATH.exists() or SCHEMA_PATH.read_text() != rendered:
print(
f"{SCHEMA_PATH.name} is out of sync with {PRICES_PATH.name}. "
f"Run `python {Path(__file__).relative_to(REPO_ROOT)}` and commit the result."
)
return 1
print(f"{SCHEMA_PATH.name} is in sync and {PRICES_PATH.name} validates against it")
return 0
if __name__ == "__main__":
sys.exit(main())

119
ci_cd/spend_logs_guard.py Normal file
View file

@ -0,0 +1,119 @@
"""Required check: a pull request that touches LiteLLM_SpendLogs in schema.prisma or a migration
needs an approving review from the spend-log owner before it can merge.
CODEOWNERS can only match file paths, and every schema change edits the same schema.prisma copies,
so a path rule would put every migration on the owner. This gate reads the diff instead: when no
added line in the watched files mentions the table it passes at once, otherwise it passes only once
the owner's latest review on the pull request is an approval. The workflow re-runs on every review.
"""
from __future__ import annotations
import argparse
import re
import subprocess
import sys
from collections.abc import Sequence
from pathlib import Path
from typing import Final
from pydantic import BaseModel, ConfigDict, TypeAdapter
WATCHED: Final = (
"schema.prisma",
"litellm/proxy/schema.prisma",
"litellm-proxy-extras/litellm_proxy_extras/schema.prisma",
)
MIGRATIONS_PREFIX: Final = "litellm-proxy-extras/litellm_proxy_extras/migrations/"
TABLE: Final = re.compile(r"(?<![A-Za-z0-9])LiteLLM_SpendLogs", re.IGNORECASE)
class Reviewer(BaseModel):
model_config = ConfigDict(frozen=True, extra="ignore")
login: str
class Review(BaseModel):
model_config = ConfigDict(frozen=True, extra="ignore")
state: str
user: Reviewer
_REVIEWS: Final = TypeAdapter(list[Review])
def watched(path: str) -> bool:
return path in WATCHED or (path.startswith(MIGRATIONS_PREFIX) and path.endswith("/migration.sql"))
def _added_lines(section: str) -> tuple[str, ...]:
body: Final = section.split("\n@@", 1)
return tuple(line[1:] for line in body[1].splitlines() if line.startswith("+")) if len(body) == 2 else ()
def spend_log_hits(diff: str) -> tuple[str, ...]:
"""`path: added line` for every added line in a watched file that names the table."""
sections: Final = re.split(r"^diff --git [^\n]*\n", diff, flags=re.MULTILINE)
return tuple(f"{path}: {line.strip()}" for path, line in _watched_added_lines(sections) if TABLE.search(line))
def _watched_added_lines(sections: Sequence[str]) -> tuple[tuple[str, str], ...]:
pairs: Final = tuple((_path_of(section), section) for section in sections)
return tuple(
(path, line) for path, section in pairs if watched(path) for line in _added_lines(section)
) # comprehension-ok: path then its lines
def _path_of(section: str) -> str:
match: Final = re.search(r"^\+\+\+ b/(.+)$", section, flags=re.MULTILINE)
return match.group(1) if match else ""
def approved_by(reviews: Sequence[Review], login: str) -> bool:
"""True when `login`'s most recent review on the pull request is an approval."""
states: Final = tuple(review.state for review in reviews if review.user.login == login)
return bool(states) and states[-1] == "APPROVED"
def added_diff(base: str, head: str) -> str:
result: Final = subprocess.run(
("git", "diff", "--unified=0", "--no-color", f"{base}..{head}", "--", *WATCHED, MIGRATIONS_PREFIX),
check=True,
capture_output=True,
text=True,
)
return result.stdout
class Arguments(BaseModel):
model_config = ConfigDict(frozen=True)
base: str
head: str
approver: str
reviews: Path
def main(argv: Sequence[str]) -> int:
parser: Final = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--base", required=True, help="merge base of the pull request")
parser.add_argument("--head", required=True, help="head sha of the pull request")
parser.add_argument("--approver", required=True, help="GitHub login whose approval unlocks the merge")
parser.add_argument("--reviews", required=True, help="JSON file with the pull request's reviews")
args: Final = Arguments.model_validate(vars(parser.parse_args(argv)))
hits: Final = spend_log_hits(added_diff(args.base, args.head))
if not hits:
sys.stdout.write("No added line in schema.prisma or a migration names LiteLLM_SpendLogs; nothing to review.\n")
return 0
if approved_by(_REVIEWS.validate_json(args.reviews.read_bytes()), args.approver):
sys.stdout.write(f"LiteLLM_SpendLogs change approved by {args.approver}: {', '.join(hits)}\n")
return 0
listed: Final = "".join(f" {hit}\n" for hit in hits)
sys.stdout.write(
f"This pull request changes LiteLLM_SpendLogs and needs an approving review from @{args.approver}:\n"
f"{listed}The check re-runs when a review is submitted.\n"
)
return 1
if __name__ == "__main__":
sys.exit(main(sys.argv[1:]))

View file

@ -0,0 +1,76 @@
# Use your ChatGPT subscription through LiteLLM
Sign into ChatGPT in Codex or Pi and send model requests through your existing LiteLLM proxy. The client keeps your login and refreshes it. Each request uses that client's ChatGPT account, while a separate LiteLLM virtual key controls gateway access and usage attribution.
An administrator adds a `chatgpt/` model using the usual model configuration. No ChatGPT login needs to be added to the proxy. This uses the same separate-key convention as provider passthrough, including Claude Code subscription forwarding.
## Add a model to the proxy
Reuse your existing proxy. If you need one, follow the [proxy quickstart](https://docs.litellm.ai/docs/proxy/docker_quick_start). Merge this entry into its model list and restart the proxy:
```yaml
model_list:
- model_name: gpt-6-luna
litellm_params:
model: chatgpt/gpt-6-luna
model_info:
mode: responses
```
Use a model available to the caller's subscription. The virtual key must allow the exposed model name. Leave saved ChatGPT logins and API credentials unset for a client-only deployment. Explicit cross-provider fallbacks can incur separate API charges; do not configure them for subscription-only traffic.
## Configure Codex
Use your existing LiteLLM URL and a virtual key assigned to your gateway user. Export that key as `LITELLM_API_KEY` in the environment that launches Codex. Keep the value out of committed configuration.
Add this provider to `~/.codex/config.toml`, preserving your other settings:
```toml
model = "gpt-6-luna"
model_provider = "litellm_subscription"
[model_providers.litellm_subscription]
name = "LiteLLM subscription"
base_url = "https://litellm.example.com/v1"
wire_api = "responses"
supports_websockets = false
requires_openai_auth = true
env_http_headers = { "x-litellm-api-key" = "LITELLM_API_KEY" }
```
Replace `https://litellm.example.com` with your proxy URL. A local proxy normally uses `http://localhost:4000/v1`. Launch Codex, sign into ChatGPT if needed, and send a prompt with a model available to your subscription. The selected model must match the name exposed by the gateway.
For a managed workspace rollout, use OpenAI's [gateway configuration guide](https://learn.chatgpt.com/docs/enterprise/sign-in-with-chatgpt-through-a-gateway).
## Native compatibility and Pi
For a native Codex model catalog and request format, change Codex's `base_url` to `https://litellm.example.com/backend-api/codex`. This route uses the same credentials but does not need a model deployment or resolve model aliases. Restricted keys must allow the upstream model name.
### Configure Pi
Sign into the built-in `openai-codex` provider in Pi. Merge this override into `~/.pi/agent/models.json` and export your LiteLLM virtual key as `LITELLM_API_KEY` before launching Pi:
```json
{
"providers": {
"openai-codex": {
"baseUrl": "https://litellm.example.com/backend-api/codex",
"headers": {
"x-litellm-api-key": "LITELLM_API_KEY"
}
}
}
}
```
Keep Pi's subscription login as the provider credential. The header value names an environment variable, not a key to paste into the file.
## Verify account usage
Send a prompt and check the completed request in LiteLLM's Logs page. Its gateway user/key and token counts should match the caller. Give each person their own LiteLLM virtual key for separate attribution. The gateway credential and ChatGPT login are independently authenticated; LiteLLM does not establish that both identities belong to the same person.
The native routes are `GET /backend-api/codex/models` and `POST /backend-api/codex/responses`. Model requests use the upstream model name, so a restricted LiteLLM key must allow that name. OpenAI's model catalog is returned in its native format; seeing a model there does not override gateway permissions.
The proxy forwards the ChatGPT bearer and account header to the fixed OpenAI Codex endpoint. It removes gateway credentials and cookies from the upstream request and redacts credential headers in logging callbacks. The client remains responsible for refresh and reauthentication. Native passthrough returns authentication and quota errors without switching subscriptions. Standard model routes retain the caller login across ChatGPT retries; explicitly configured cross-provider fallbacks use their own credentials. An explicit workspace header is preserved for OpenAI to validate.
This flow does not connect accounts through the dashboard or pool subscription allowances. Use a Codex build that supports ChatGPT authentication with a custom model provider. The integration forwards HTTP and SSE requests; it does not add native WebSocket or other Codex backend routes.

View file

@ -0,0 +1,41 @@
# Bedrock batch endpoints
Set `aws_bedrock_control_plane_endpoint` on a deployment when batch create, retrieve, and cancel requests must use a custom Bedrock control-plane endpoint. The proxy signs these requests for the configured URL. Without this setting, batch requests use the standard regional Bedrock endpoint
```yaml
model_list:
- model_name: bedrock-batch
litellm_params:
model: bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0
aws_region_name: us-east-1
aws_access_key_id: os.environ/AWS_ACCESS_KEY_ID
aws_secret_access_key: os.environ/AWS_SECRET_ACCESS_KEY
aws_batch_role_arn: os.environ/AWS_BATCH_ROLE_ARN
s3_output_bucket_name: os.environ/AWS_S3_OUTPUT_BUCKET_NAME
aws_bedrock_control_plane_endpoint: os.environ/BEDROCK_CONTROL_PLANE_ENDPOINT
```
Set `BEDROCK_CONTROL_PLANE_ENDPOINT` to your Bedrock control-plane VPC endpoint or compatible gateway base URL, without `/model-invocation-job`. The proxy adds the operation path. The endpoint must be reachable from the proxy, and the configured AWS region must match the endpoint's signing region
`api_base` and `aws_bedrock_runtime_endpoint` configure runtime inference separately. They do not override the batch control-plane endpoint. `s3_endpoint_url` configures S3 file access separately
The control-plane setting belongs in deployment configuration. Sending `aws_bedrock_control_plane_endpoint` in an API request body is rejected. There is no automatic `AWS_BEDROCK_CONTROL_PLANE_ENDPOINT` environment fallback, so reference the desired environment variable explicitly as shown above
Start the proxy with this configuration and set `LITELLM_API_BASE` to its URL. Set `BEDROCK_BATCH_INPUT_URI` to a valid input JSONL file in S3 that the batch service role can read
```bash
curl -sS "$LITELLM_API_BASE/v1/batches" \
-H "Authorization: Bearer $LITELLM_MASTER_KEY" \
-H 'Content-Type: application/json' \
-d "{\"model\":\"bedrock-batch\",\"input_file_id\":\"$BEDROCK_BATCH_INPUT_URI\",\"endpoint\":\"/v1/chat/completions\",\"completion_window\":\"24h\"}"
```
Use the returned `id` as `BATCH_ID` to retrieve or cancel the same job through the deployment's configured endpoint
```bash
curl -sS "$LITELLM_API_BASE/v1/batches/$BATCH_ID?model=bedrock-batch" \
-H "Authorization: Bearer $LITELLM_MASTER_KEY"
curl -sS -X POST "$LITELLM_API_BASE/v1/batches/$BATCH_ID/cancel?model=bedrock-batch" \
-H "Authorization: Bearer $LITELLM_MASTER_KEY"
```

View file

@ -6563,6 +6563,87 @@
],
"title": "litellm_project_model_rate_limit_used_metric",
"type": "timeseries"
},
{
"collapsed": false,
"gridPos": {
"h": 1,
"w": 24,
"x": 0,
"y": 447
},
"id": 115,
"panels": [],
"title": "Agent tracing export (needs LITELLM_LENS_URL)",
"type": "row"
},
{
"datasource": {
"type": "prometheus",
"uid": "${DS_PROMETHEUS}"
},
"description": "Request records the Lens exporter delivered to the Lens service against the records it dropped before the service accepted them",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"drawStyle": "line",
"fillOpacity": 10,
"lineWidth": 1,
"showPoints": "never",
"spanNulls": false
},
"unit": "short"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 24,
"x": 0,
"y": 448
},
"id": 116,
"options": {
"legend": {
"calcs": [],
"displayMode": "list",
"placement": "bottom",
"showLegend": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"type": "prometheus",
"uid": "${DS_PROMETHEUS}"
},
"editorMode": "code",
"expr": "sum(rate(litellm_lens_export_rows_written_total[$__rate_interval]))",
"legendFormat": "written",
"range": true,
"refId": "A"
},
{
"datasource": {
"type": "prometheus",
"uid": "${DS_PROMETHEUS}"
},
"editorMode": "code",
"expr": "sum(rate(litellm_lens_export_rows_dropped_total[$__rate_interval]))",
"legendFormat": "dropped",
"range": true,
"refId": "B"
}
],
"title": "litellm_lens_export_rows_written and litellm_lens_export_rows_dropped rate",
"type": "timeseries"
}
],
"preload": false,

View file

@ -1,10 +1,10 @@
# LiteLLM All Prometheus Metrics dashboard
Every `litellm_*` metric family the proxy can expose on `/metrics` (141 families across 97 panels), grouped into rows: proxy traffic, latency, spend and tokens, cache, LLM API deployments, key and team rate limits, budgets, guardrails, MCP, managed files and batches, users and teams, the Redis circuit breaker, the spend log cleanup job, and the `prometheus_system` service callback metrics (per-service latency, request and failure rates, spend update queue sizes). Panel titles are the metric names so you can grep the JSON for the metric you care about
Every `litellm_*` metric family the proxy can expose on `/metrics` (143 families across 98 panels), grouped into rows: proxy traffic, latency, spend and tokens, cache, LLM API deployments, key and team rate limits, budgets, guardrails, MCP, managed files and batches, users and teams, the Redis circuit breaker, the spend log cleanup job, the `prometheus_system` service callback metrics (per-service latency, request and failure rates, spend update queue sizes), project model rate limits, and the agent tracing export to Lens. Panel titles are the metric names so you can grep the JSON for the metric you care about
Import `grafana_dashboard.json` from **Dashboards > New > Import** and pick your Prometheus data source when prompted (the `DS_PROMETHEUS` variable). Counters are plotted as `rate()` over `$__rate_interval`, histograms as p50 / p95 / p99, gauges as the raw value grouped by the most useful label. Every query names the metric exactly as the proxy emits it (counters carry the `_total` suffix the Prometheus client adds), and `tests/unit/integrations/test_prometheus_metric_name_consistency.py` fails if a metric is renamed without updating this dashboard
The first eleven rows need only `callbacks: ["prometheus"]`. The last three rows and the `litellm_admission_*` panels are emitted by other subsystems and stay empty until those are on: the service callback row needs `service_callback: ["prometheus_system"]` in `litellm_settings`, the circuit breaker row needs a Redis cache, the cleanup row needs spend log retention, and admission control needs its middleware enabled. Within the base rows, many panels only fill in once the matching feature is in use: budgets need keys, teams, users or orgs with `max_budget` set, cache panels need caching on, guardrail and MCP panels need those features configured, deployment health needs the router with more than one deployment or a failure to record, and `litellm_in_flight_requests` needs traffic at scrape time. An empty panel for a feature you do not use is expected
The first eleven rows and the project model rate limits row need only `callbacks: ["prometheus"]`. The other rows and the `litellm_admission_*` panels are emitted by other subsystems and stay empty until those are on: the service callback row needs `service_callback: ["prometheus_system"]` in `litellm_settings`, the circuit breaker row needs a Redis cache, the cleanup row needs spend log retention, the agent tracing export row needs `LITELLM_LENS_URL`, and admission control needs its middleware enabled. Within the base rows, many panels only fill in once the matching feature is in use: budgets need keys, teams, users or orgs with `max_budget` set, cache panels need caching on, guardrail and MCP panels need those features configured, deployment health needs the router with more than one deployment or a failure to record, and `litellm_in_flight_requests` needs traffic at scrape time. An empty panel for a feature you do not use is expected
## Pre-requisites

View file

@ -1,4 +1,4 @@
-----BEGIN PUBLIC KEY-----
MFkwEwYHKoZIzj0CAQYIKoZIzj0DAQcDQgAEKi4ivqGpE231OGH50PKbqy1Y1Kkb
POJC8+i2Wko82gBOUCe3M0Vw86H/4rhUhfoYEti4gdJ9wZbYmK0I2EE96g==
MFkwEwYHKoZIzj0CAQYIKoZIzj0DAQcDQgAEG06dpKM+vEDnmp2zcjyXgVCXxzRg
JA5H2LnmwBKQafZGka6OOrWujCEWATQewiAgyD4WQdJ0pvU9RC4mvNAjvQ==
-----END PUBLIC KEY-----

View file

@ -0,0 +1,201 @@
import argparse
import os
import sys
from collections.abc import Iterator, Sequence
from dataclasses import dataclass
from itertools import accumulate, repeat, takewhile
from typing import Final, Literal, TextIO
import httpx
from pydantic import BaseModel, ConfigDict, JsonValue, ValidationError
from litellm.vector_stores.utils import is_litellm_written_metadata_key
DEFAULT_API_BASE: Final = "https://api.openai.com/v1"
LIST_PAGE_SIZE: Final = "100"
class VectorStore(BaseModel):
model_config = ConfigDict(frozen=True, extra="ignore")
id: str
metadata: dict[str, JsonValue] | None = None
class VectorStorePage(BaseModel):
model_config = ConfigDict(frozen=True, extra="ignore")
data: tuple[VectorStore, ...]
has_more: bool = False
class OpenAIErrorBody(BaseModel):
model_config = ConfigDict(frozen=True, extra="ignore")
message: str = ""
class OpenAIErrorResponse(BaseModel):
model_config = ConfigDict(frozen=True, extra="ignore")
error: OpenAIErrorBody
@dataclass(frozen=True, slots=True)
class Options:
api_key: str
api_base: str
vector_store_ids: tuple[str, ...]
extra_keys: frozenset[str]
apply: bool
@dataclass(frozen=True, slots=True)
class StoreResult:
store_id: str
matched_keys: tuple[str, ...]
status: Literal["dry_run", "updated", "failed"]
error: str = ""
def is_proxy_written_key(key: str, extra_keys: frozenset[str]) -> bool:
return is_litellm_written_metadata_key(key) or key in extra_keys
def matching_keys(store: VectorStore, extra_keys: frozenset[str]) -> tuple[str, ...]:
return tuple(sorted(key for key in (store.metadata or {}) if is_proxy_written_key(key, extra_keys)))
def fetch_page(client: httpx.Client, after: str | None) -> VectorStorePage:
params: Final = {"limit": LIST_PAGE_SIZE} if after is None else {"limit": LIST_PAGE_SIZE, "after": after}
return VectorStorePage.model_validate_json(client.get("/vector_stores", params=params).raise_for_status().content)
def fetch_next_page(client: httpx.Client, page: VectorStorePage | None) -> VectorStorePage | None:
if page is None or not page.has_more or not page.data:
return None
return fetch_page(client, after=page.data[-1].id)
def list_all_stores(client: httpx.Client) -> Iterator[VectorStore]:
pages: Final = accumulate(
repeat(None), lambda page, _: fetch_next_page(client, page), initial=fetch_page(client, after=None)
)
for page in takewhile(lambda page: page is not None, pages):
yield from page.data
def retrieve_stores(client: httpx.Client, store_ids: tuple[str, ...]) -> Iterator[VectorStore]:
for store_id in store_ids:
yield VectorStore.model_validate_json(client.get(f"/vector_stores/{store_id}").raise_for_status().content)
def describe_http_error(response: httpx.Response) -> str:
try:
message: Final = OpenAIErrorResponse.model_validate_json(response.content).error.message
except ValidationError:
return f"HTTP {response.status_code}"
return f"HTTP {response.status_code}: {message}"
def delete_keys(client: httpx.Client, store_id: str, keys: tuple[str, ...]) -> StoreResult:
try:
response: Final = client.post(f"/vector_stores/{store_id}", json={"metadata": dict.fromkeys(keys)})
except httpx.TransportError as error:
return StoreResult(store_id=store_id, matched_keys=keys, status="failed", error=type(error).__name__)
if response.is_success:
return StoreResult(store_id=store_id, matched_keys=keys, status="updated")
return StoreResult(store_id=store_id, matched_keys=keys, status="failed", error=describe_http_error(response))
def clean_store(client: httpx.Client, store: VectorStore, options: Options) -> StoreResult | None:
keys: Final = matching_keys(store, options.extra_keys)
if not keys:
return None
if not options.apply:
return StoreResult(store_id=store.id, matched_keys=keys, status="dry_run")
return delete_keys(client, store.id, keys)
def format_result(result: StoreResult) -> str:
key_list: Final = ", ".join(result.matched_keys)
count: Final = len(result.matched_keys)
match result.status:
case "dry_run":
return f"{result.store_id}: would delete {count} keys: {key_list}"
case "updated":
return f"{result.store_id}: deleted {count} keys: {key_list}"
case "failed":
return f"{result.store_id}: failed to delete {count} keys ({result.error}): {key_list}"
def clean_and_report(
client: httpx.Client, stores: Iterator[VectorStore], options: Options, out: TextIO
) -> Iterator[StoreResult | None]:
for store in stores:
result: Final = clean_store(client, store, options)
if result is not None:
print(format_result(result), file=out)
yield result
def run(options: Options, client: httpx.Client, out: TextIO) -> int:
stores: Final = (
retrieve_stores(client, options.vector_store_ids) if options.vector_store_ids else list_all_stores(client)
)
scanned: Final = tuple(clean_and_report(client, stores, options, out))
results: Final = tuple(result for result in scanned if result is not None)
updated: Final = sum(1 for result in results if result.status == "updated")
failed: Final = sum(1 for result in results if result.status == "failed")
mode: Final = "applied" if options.apply else "dry run, pass --apply to delete"
print(
f"scanned={len(scanned)} matched={len(results)} updated={updated} failed={failed} ({mode})",
file=out,
)
return 1 if failed else 0
def parse_options(argv: Sequence[str] | None) -> Options:
parser: Final = argparse.ArgumentParser(
description="Remove LiteLLM proxy-written keys from OpenAI vector store metadata. Dry run by default."
)
parser.add_argument("--api-key", default=os.environ.get("OPENAI_API_KEY"), help="defaults to $OPENAI_API_KEY")
parser.add_argument("--api-base", default=DEFAULT_API_BASE)
parser.add_argument("--vector-store-id", action="append", default=[], help="repeatable, defaults to every store")
parser.add_argument("--extra-key", action="append", default=[], help="repeatable, another key name to delete")
parser.add_argument("--apply", action="store_true", help="delete the matching keys instead of listing them")
args: Final = parser.parse_args(argv)
if not args.api_key:
parser.error("--api-key or OPENAI_API_KEY is required")
return Options(
api_key=args.api_key,
api_base=args.api_base.rstrip("/"),
vector_store_ids=tuple(args.vector_store_id),
extra_keys=frozenset(args.extra_key),
apply=args.apply,
)
def main(argv: Sequence[str] | None = None, transport: httpx.BaseTransport | None = None) -> int:
options: Final = parse_options(argv)
with httpx.Client(
base_url=options.api_base,
headers={"Authorization": f"Bearer {options.api_key}"},
timeout=60.0,
transport=transport,
) as client:
try:
return run(options, client, sys.stdout)
except httpx.HTTPStatusError as error:
sys.stderr.write(f"could not read vector stores: {describe_http_error(error.response)}\n")
return 2
except httpx.TransportError as error:
sys.stderr.write(f"could not reach {options.api_base}: {type(error).__name__}\n")
return 2
except ValidationError:
sys.stderr.write(f"unexpected vector store response shape from {options.api_base}\n")
return 2
if __name__ == "__main__":
sys.exit(main())

View file

@ -1,10 +1,10 @@
# syntax=docker/dockerfile:1.7
# Base image for building
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
# Runtime image
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Pinned by digest like the other base images; bump explicitly on Node upgrades.
ARG UI_BUILD_IMAGE=node:24.19-alpine3.24@sha256:d32cdf619f63fe0471182d08996dd516c6275bb5fd31ae06e55a570bd9e1ad43
@ -59,6 +59,7 @@ RUN apk add --no-cache \
python-3.13 \
python-3.13-dev \
openssl \
libssl3 \
openssl-dev \
nodejs \
npm \
@ -122,7 +123,7 @@ ENV LITELLM_RELEASE_TAG=${LITELLM_RELEASE_TAG}
USER root
# node (without npm) is required by the prisma CLI at runtime
RUN apk add --no-cache bash openssl tzdata nodejs python-3.13 libsndfile libevent
RUN apk add --no-cache bash openssl libssl3 tzdata nodejs python-3.13 libsndfile libevent
COPY --from=pgbouncer-builder /usr/local/bin/pgbouncer /usr/local/bin/pgbouncer
WORKDIR /app

View file

@ -1,8 +1,8 @@
# syntax=docker/dockerfile:1.7
# Base images
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Pinned by digest like the other base images; bump explicitly on Node upgrades.
ARG UI_BUILD_IMAGE=node:24.19-alpine3.24@sha256:d32cdf619f63fe0471182d08996dd516c6275bb5fd31ae06e55a570bd9e1ad43
@ -60,6 +60,7 @@ RUN for i in 1 2 3; do \
coreutils \
curl \
openssl \
libssl3 \
libsndfile \
nodejs \
npm && break || sleep 5; \
@ -134,7 +135,7 @@ RUN for i in 1 2 3; do \
apk upgrade --no-cache && break || sleep 5; \
done && \
for i in 1 2 3; do \
apk add --no-cache python-3.13 bash openssl tzdata libsndfile nodejs libevent && break || sleep 5; \
apk add --no-cache python-3.13 bash openssl libssl3 tzdata libsndfile nodejs libevent && break || sleep 5; \
done
COPY --from=pgbouncer-builder /usr/local/bin/pgbouncer /usr/local/bin/pgbouncer

View file

@ -30,7 +30,7 @@ RUN uv venv --python python && \
"google-cloud-aiplatform==1.133.0" \
"google-genai==1.37.0" \
"anthropic[vertex]==0.84.0" \
"grpcio==1.78.0" \
"grpcio==1.84.1" \
"prometheus-client==0.20.0" \
"langfuse==2.59.7" \
"opentelemetry-api==1.28.0" \

View file

@ -11,7 +11,7 @@
# beyond local evaluation, pin the image to a specific release tag.
services:
litellm:
image: docker.litellm.ai/berriai/litellm:main-stable
image: docker.litellm.ai/berriai/litellm:latest
ports:
# LITELLM_BIND is empty by default, so this stays "4000:4000". The quickstart
# script sets it to "127.0.0.1:" so new installs listen on this machine only.

View file

@ -15,9 +15,10 @@ sys.path.insert(
) # Adds the parent directory to the system path
import json
import sys
from typing import Any, List, Literal, Optional
from typing import Final, List, Literal, Optional
from fastapi import HTTPException
from pydantic import TypeAdapter
import litellm
from litellm._logging import verbose_proxy_logger
@ -29,11 +30,13 @@ from litellm.llms.custom_httpx.http_handler import (
get_async_httpx_client,
httpxSpecialProvider,
)
from litellm.llms.openai.common_utils import decoded_json_object
from litellm.proxy._types import UserAPIKeyAuth
from litellm.proxy.guardrails.guardrail_helpers import should_proceed_based_on_metadata
from litellm.types.guardrails import GuardrailEventHooks
GUARDRAIL_NAME = "aporia"
_DECODED_JSON: Final = TypeAdapter(object)
class AporiaGuardrail(CustomGuardrail):
@ -68,7 +71,7 @@ class AporiaGuardrail(CustomGuardrail):
async def prepare_aporia_request(
self, new_messages: List[dict], response_string: Optional[str] = None
) -> dict:
data: dict[str, Any] = {}
data: dict[str, object] = {}
if new_messages is not None:
data["messages"] = new_messages
if response_string is not None:
@ -122,8 +125,10 @@ class AporiaGuardrail(CustomGuardrail):
verbose_proxy_logger.debug("Aporia AI response: %s", response.text)
if response.status_code == 200:
# check if the response was flagged
_json_response = response.json()
action: str = _json_response.get(
_json_response: Final = decoded_json_object(
_DECODED_JSON.validate_python(response.json())
)
action: Final = _json_response.get(
"action"
) # possible values are modify, passthrough, block, rephrase
if action == "block":

View file

@ -7,9 +7,10 @@
## This accepts a list of user id's for whom calls will be rejected
from typing import Literal, Optional
from typing import Final, Literal, Optional
from fastapi import HTTPException
from pydantic import TypeAdapter
import litellm
from litellm._internal_context import with_service_target
@ -19,6 +20,9 @@ from litellm.integrations.custom_logger import CustomLogger
from litellm.proxy._types import LiteLLM_EndUserTable, UserAPIKeyAuth
from litellm.proxy.common_utils.user_api_key_cache import AUTH_OBJECTS_TARGET
from litellm.proxy.utils import PrismaClient
from litellm.repositories.table_repositories import EndUserRepository
_REQUEST_BODY_VALUE: Final = TypeAdapter(object)
class ENTERPRISE_BlockedUserList(CustomLogger):
@ -94,15 +98,14 @@ class ENTERPRISE_BlockedUserList(CustomLogger):
)
if end_user_cache_obj is None and self.prisma_client is not None:
# check db
end_user_obj = (
await self.prisma_client.db.litellm_endusertable.find_unique(
where={"user_id": user}
)
)
end_user_id: Final = _REQUEST_BODY_VALUE.validate_python(user)
end_user_obj = await EndUserRepository(
self.prisma_client
).table.find_unique(where={"user_id": end_user_id})
if end_user_obj is None: # user not in db - assume not blocked
end_user_obj = LiteLLM_EndUserTable(user_id=user, blocked=False)
cache.set_cache(key=cache_key, value=end_user_obj, ttl=60)
if end_user_obj is not None and end_user_obj.blocked is True:
if end_user_obj.blocked is True:
raise HTTPException(
status_code=400,
detail={

View file

@ -6,7 +6,10 @@
# +-----------------------------------------------+
# Thank you users! We ❤️ you! - Krrish & Ishaan
from typing import Final
from fastapi import HTTPException
from pydantic import TypeAdapter
import litellm
from litellm._logging import verbose_proxy_logger
@ -15,6 +18,8 @@ from litellm.proxy._types import UserAPIKeyAuth
from litellm.proxy.guardrails._content_utils import iter_message_text
from litellm.types.utils import CallTypesLiteral
_THRESHOLD_VALUE: Final = TypeAdapter(object)
class ENTERPRISE_GoogleTextModeration(CustomLogger):
user_api_key_cache = None
@ -69,9 +74,11 @@ class ENTERPRISE_GoogleTextModeration(CustomLogger):
f"{category}_confidence_threshold",
default_confidence_threshold,
)
set_confidence_value = getattr(
self,
f"{category}_confidence_threshold",
set_confidence_value = _THRESHOLD_VALUE.validate_python(
getattr(
self,
f"{category}_confidence_threshold",
)
)
verbose_proxy_logger.info(
f"Google Text Moderation: {category}_confidence_threshold: {set_confidence_value}"

View file

@ -1,15 +1,20 @@
# this is an example endpoint to receive data from litellm
from typing import Final
from fastapi import FastAPI, HTTPException, Request
from pydantic import TypeAdapter
app = FastAPI()
_LOGGED_EVENT: Final = TypeAdapter(object)
@app.post("/log-event")
async def log_event(request: Request):
try:
print("Received /log-event request") # noqa
# Assuming the incoming request has JSON data
data = await request.json()
data: Final = _LOGGED_EVENT.validate_python(await request.json())
print("Received request data:") # noqa
print(data) # noqa

View file

@ -8,25 +8,30 @@
## This provides an LLM Guard Integration for content moderation on the proxy
import asyncio
from typing import Final, Optional
from collections.abc import Mapping
from typing import Final
import aiohttp
from fastapi import HTTPException
from pydantic import TypeAdapter
import litellm
from litellm._logging import verbose_proxy_logger
from litellm.integrations.custom_logger import CustomLogger
from litellm.llms.openai.common_utils import decoded_json_object
from litellm.proxy._types import UserAPIKeyAuth
from litellm.secret_managers.main import get_secret_str
from litellm.types.utils import CallTypesLiteral
_DECODED_JSON: Final = TypeAdapter(object)
class _ENTERPRISE_LLMGuard(CustomLogger):
# Class variables or attributes
def __init__(
self,
mock_testing: bool = False,
mock_redacted_text: Optional[dict] = None,
mock_redacted_text: Mapping[str, object] | None = None,
):
self.mock_redacted_text = mock_redacted_text
self.llm_guard_mode = litellm.llm_guard_mode
@ -66,7 +71,9 @@ class _ENTERPRISE_LLMGuard(CustomLogger):
async with session.post(
analyze_url, json={"prompt": text}
) as response:
redacted_text = await response.json()
redacted_text = _DECODED_JSON.validate_python(
await response.json()
)
verbose_proxy_logger.debug(
f"LLM Guard: Received response - {redacted_text}"
)
@ -77,12 +84,13 @@ class _ENTERPRISE_LLMGuard(CustomLogger):
"error": f"Invalid content moderation response: {redacted_text}"
},
)
if redacted_text.get("is_valid", None) is False:
redacted_response: Final = decoded_json_object(redacted_text)
if redacted_response.get("is_valid", None) is False:
raise HTTPException(
status_code=400,
detail={"error": "Violated content safety policy"},
)
sanitized_prompt = redacted_text.get("sanitized_prompt")
sanitized_prompt: Final = redacted_response.get("sanitized_prompt")
return sanitized_prompt if isinstance(sanitized_prompt, str) else text
except Exception as e:
verbose_proxy_logger.exception(

View file

@ -20,9 +20,15 @@ class EnterpriseCustomGuardrailHelper:
the tag's value (the mode for that tag).
- If no request tag matches, fall back to default mode(s).
"""
from litellm.litellm_core_utils.core_helpers import (
get_metadata_variable_name_from_kwargs,
)
from litellm.litellm_core_utils.litellm_logging import (
StandardLoggingPayloadSetup,
)
from litellm.router_strategy.tag_based_routing import (
without_baggage_mapped_tags,
)
from litellm.proxy._types import CommonProxyErrors
from litellm.proxy.proxy_server import premium_user
@ -36,9 +42,13 @@ class EnterpriseCustomGuardrailHelper:
proxy_server_request = data.get("proxy_server_request", {})
request_tags = StandardLoggingPayloadSetup.get_request_tags(
litellm_params=data,
proxy_server_request=proxy_server_request,
request_tags = without_baggage_mapped_tags(
StandardLoggingPayloadSetup.get_request_tags(
litellm_params=data,
proxy_server_request=proxy_server_request,
),
data.get(get_metadata_variable_name_from_kwargs(data)),
keep_serving_deployment_tags=True,
)
# Check if any request tag matches a Mode tag key

View file

@ -2,6 +2,8 @@ import os
from fastapi import HTTPException, status
from litellm.proxy._types import MODEL_LIST_ROUTES
class EnterpriseRouteChecks:
@staticmethod
@ -43,7 +45,7 @@ class EnterpriseRouteChecks:
# Routes that should remain accessible even when LLM API endpoints are disabled.
# These are read-only model listing routes needed by the Admin UI.
LLM_API_EXEMPT_ROUTES = ["/models", "/v1/models"]
LLM_API_EXEMPT_ROUTES = MODEL_LIST_ROUTES
@staticmethod
def should_call_route(route: str):

View file

@ -1,14 +1,16 @@
from typing import Any, Optional
from collections.abc import Awaitable, Callable
from fastapi import Request
from litellm._logging import verbose_proxy_logger
from litellm.proxy._types import ProxyException, UserAPIKeyAuth
from litellm.proxy._types import ProxyException
async def enterprise_custom_auth(
request: Request, api_key: str, user_custom_auth: Optional[Any]
) -> Optional[UserAPIKeyAuth]:
request: Request,
api_key: str,
user_custom_auth: Callable[[Request, str], Awaitable[object]] | None,
) -> object:
from litellm_enterprise.proxy.proxy_server import custom_auth_settings
if user_custom_auth is None:

View file

@ -2,27 +2,37 @@
Polls LiteLLM_ManagedObjectTable to check if the batch job is complete, and if the cost has been tracked.
"""
from collections.abc import Sequence
import asyncio
import json
import re
from collections.abc import Mapping, Sequence
from dataclasses import replace as dataclasses_replace
from datetime import datetime, timedelta, timezone
from types import MappingProxyType
from typing import TYPE_CHECKING, Final, List, Literal, Optional, Protocol, Tuple, cast
from litellm_enterprise.proxy.common_utils.managed_object_attribution import (
ManagedObjectCreatorAttribution,
ManagedObjectRow,
row_metadata,
)
import litellm
from litellm._logging import verbose_proxy_logger
from litellm._uuid import uuid
from litellm.constants import (
CLI_SESSION_KEY_PREFIX,
BATCH_LINE_ITEMS_MAX_PART_BYTES,
MANAGED_OBJECT_STALENESS_CUTOFF_DAYS,
MAX_OBJECTS_PER_POLL_CYCLE,
)
from litellm.repositories.table_repositories import ManagedObjectRepository
from litellm.repositories.team_repository import TeamRepository
from litellm.repositories.user_repository import UserRepository
from litellm.repositories.verification_token_repository import VerificationTokenRepository
if TYPE_CHECKING:
from prisma import models as prisma_models
from litellm.batches.batch_line_items import BatchFileProvider
from litellm.integrations.prometheus import PrometheusLogger
from litellm.proxy.utils import PrismaClient, ProxyLogging
from litellm.repositories.prisma_protocols import TableActions
@ -31,7 +41,12 @@ if TYPE_CHECKING:
from litellm.types.utils import LiteLLMBatch
def _error_names_resource(error: Exception, resource_id: str) -> bool:
return resource_id in re.findall(r"""[^\s"'`]+""", str(error))
CHECK_BATCH_COST_USER_AGENT = "LiteLLM Proxy/CheckBatchCost"
RECENT_BATCH_POLL_WINDOW: Final = timedelta(hours=48)
PROVIDER_TERMINAL_BATCH_STATUSES: Final[Tuple[str, ...]] = (
"completed",
@ -47,29 +62,6 @@ TERMINAL_MANAGED_OBJECT_STATUSES: Final[Tuple[str, ...]] = (
)
class _ManagedObjectRow(Protocol):
@property
def id(self) -> str: ...
@property
def unified_object_id(self) -> str: ...
@property
def created_by(self) -> str | None: ...
@property
def file_object(self) -> object: ...
@property
def org_id(self) -> str | None: ...
@property
def api_key(self) -> str | None: ...
@property
def team_id(self) -> str | None: ...
class _ReadableFileContent(Protocol):
async def read(self) -> bytes: ...
@ -87,14 +79,10 @@ async def _file_content_bytes(file_content: object) -> bytes:
return cast(bytes, file_content)
def _managed_object_table(prisma_client: "PrismaClient") -> "TableActions[_ManagedObjectRow]":
def _managed_object_table(prisma_client: "PrismaClient") -> "TableActions[ManagedObjectRow]":
return ManagedObjectRepository(prisma_client).table
def _user_table(prisma_client: "PrismaClient") -> "TableActions[prisma_models.LiteLLM_UserTable]":
return UserRepository(prisma_client).table
def _token_table(prisma_client: "PrismaClient") -> "TableActions[prisma_models.LiteLLM_VerificationToken]":
return VerificationTokenRepository(prisma_client).table
@ -118,10 +106,20 @@ class CheckBatchCost:
self.prisma_client: PrismaClient = prisma_client
self.llm_router: Router = llm_router
self._track_unmanaged_batch_cost = track_unmanaged_batch_cost
self._creator_attribution: Final = ManagedObjectCreatorAttribution(prisma_client, "CheckBatchCost")
# Cached after the first poll cycle. Once we know the column is absent we skip
# the guaranteed-failing primary query on every subsequent cycle.
self._has_batch_processed_column: bool = True
self.batch_processed_support_confirmed: bool = False
self._poll_cursor: tuple[datetime, str] | None = None
self._poll_before: datetime = datetime.now(timezone.utc)
self._arrival_cursor: tuple[datetime, str] | None = None
self._arrival_before: datetime = self._poll_before
self._poll_arrivals: bool = False
self._line_item_stores: Final[set[asyncio.Task[None]]] = set()
self._line_item_store_turn: Final = asyncio.Lock()
self._poll_idle: Final = asyncio.Event()
self._poll_idle.set()
@staticmethod
def _is_missing_batch_processed_column_error(err: Exception) -> bool:
@ -149,133 +147,6 @@ class CheckBatchCost:
return
self.batch_processed_support_confirmed = True
async def _get_user_info(self, batch_id: str, user_id: Optional[str]) -> dict[str, str | None]:
"""
Look up user email and key alias by user_id for enriching the S3 callback metadata.
Returns a dict with user_api_key_user_email and user_api_key_alias (both may be None).
Returns an empty dict when user_id is None: batches created by a team or service
account key carry no user id, and find_unique(where={"user_id": None}) raises.
"""
if not user_id:
return {}
try:
user_row: prisma_models.LiteLLM_UserTable | None = await _user_table(self.prisma_client).find_unique(
where={"user_id": user_id}
)
if user_row is None:
return {}
return {
"user_api_key_user_email": getattr(user_row, "user_email", None),
"user_api_key_alias": getattr(user_row, "user_alias", None),
}
except Exception as e:
verbose_proxy_logger.error(f"CheckBatchCost: could not look up user {user_id} for batch {batch_id}: {e}")
return {}
async def _get_key_alias(self, batch_id: str, api_key: str | None, created_by: str | None) -> str | None:
"""Resolve the creating virtual key's alias from its hashed token."""
if not api_key:
return None
if created_by and api_key == f"{CLI_SESSION_KEY_PREFIX}-{created_by}":
return api_key
try:
key_row: prisma_models.LiteLLM_VerificationToken | None = await _token_table(
self.prisma_client
).find_unique(where={"token": api_key})
return getattr(key_row, "key_alias", None) if key_row is not None else None
except Exception as e:
verbose_proxy_logger.error(f"CheckBatchCost: could not look up key alias for batch {batch_id}: {e}")
return None
async def _get_team_alias(self, team_id: str | None) -> str | None:
"""Resolve a team's alias from its id."""
if not team_id:
return None
try:
team_row: prisma_models.LiteLLM_TeamTable | None = await _team_table(self.prisma_client).find_unique(
where={"team_id": team_id}
)
return getattr(team_row, "team_alias", None) if team_row is not None else None
except Exception as e:
verbose_proxy_logger.error(f"CheckBatchCost: could not look up team alias for team {team_id}: {e}")
return None
async def _get_org_id(self, job: "_ManagedObjectRow", batch_id: str) -> str | None:
org_id: Final = getattr(job, "org_id", None)
if org_id:
return org_id
api_key: Final = getattr(job, "api_key", None)
team_id: Final = getattr(job, "team_id", None)
if api_key:
try:
key_row: prisma_models.LiteLLM_VerificationToken | None = await _token_table(
self.prisma_client
).find_unique(where={"token": api_key})
key_org_id: Final = (
cast(str | None, getattr(key_row, "organization_id", None))
if key_row is not None
else None
)
if key_org_id:
return key_org_id
except Exception as e:
verbose_proxy_logger.error(
f"CheckBatchCost: could not resolve the key's org for batch {batch_id}, "
f"still trying the team's: {e}"
)
if not team_id:
return None
try:
team_row: prisma_models.LiteLLM_TeamTable | None = await _team_table(self.prisma_client).find_unique(
where={"team_id": team_id}
)
return cast(str | None, getattr(team_row, "organization_id", None)) if team_row is not None else None
except Exception as e:
verbose_proxy_logger.error(f"CheckBatchCost: could not resolve the team's org for batch {batch_id}: {e}")
return None
async def _build_creator_attribution_metadata(
self, job: "_ManagedObjectRow", batch_id: str
) -> dict[str, object]:
"""
Rebuild the spend-tracking metadata for the key, team, and tags that created the
batch so the batch-cost spend log is attributed the same way a non-batch request
is. Rows created before api_key and request_tags were persisted carry only
created_by and team_id, and fall back to those. A named creating key owns
user_api_key_alias; when it has no alias, or the key has since been rotated or
deleted, the field keeps the creating user's alias that _get_user_info filled in,
because a resolvable name is more useful on the spend row than a null.
user_api_key_org_id must be resolved here too: the spend update writer reads it
off this metadata to increment organization spend, so leaving it out silently
drops batch cost from org accounting for keys and teams that belong to one.
"""
api_key = getattr(job, "api_key", None)
team_id = getattr(job, "team_id", None)
request_tags = getattr(job, "request_tags", None)
metadata: dict[str, object] = {
"user_api_key_user_id": job.created_by,
"user_api_key": api_key,
"user_api_key_hash": api_key,
"user_api_key_team_id": team_id,
**(await self._get_user_info(batch_id, job.created_by)),
}
key_alias = await self._get_key_alias(batch_id, api_key, job.created_by)
if key_alias is not None:
metadata["user_api_key_alias"] = key_alias
team_alias = await self._get_team_alias(team_id)
if team_alias is not None:
metadata["user_api_key_team_alias"] = team_alias
org_id: Final = await self._get_org_id(job, batch_id)
if org_id is not None:
metadata["user_api_key_org_id"] = org_id
if isinstance(request_tags, list) and request_tags:
metadata["tags"] = [tag for tag in request_tags if isinstance(tag, str)]
return metadata
async def _cleanup_stale_managed_objects(self) -> None:
"""
Mark managed objects older than MANAGED_OBJECT_STALENESS_CUTOFF_DAYS days
@ -297,29 +168,9 @@ class CheckBatchCost:
f"(older than {MANAGED_OBJECT_STALENESS_CUTOFF_DAYS} days) as stale_expired"
)
if not self._has_batch_processed_column:
return
# A row already in a terminal status is never rewritten by the sweep above, so
# without this it keeps a poll-page slot forever and starves newer batches.
retired: Final = await _managed_object_table(self.prisma_client).update_many(
where={
"file_purpose": "batch",
"batch_processed": False,
"status": {"in": ["complete", "completed"]},
"created_at": {"lt": cutoff},
},
data={"batch_processed": True},
)
if retired > 0:
verbose_proxy_logger.warning(
f"CheckBatchCost: gave up on {retired} completed managed objects older than "
f"{MANAGED_OBJECT_STALENESS_CUTOFF_DAYS} days that were never costed"
)
async def _fallback_find_jobs(self) -> "Sequence[_ManagedObjectRow]":
async def _fallback_find_jobs(self) -> "Sequence[ManagedObjectRow]":
"""Query batch jobs without the batch_processed filter (for older schemas)."""
return await _managed_object_table(self.prisma_client).find_many(
return await self._find_jobs(
where={
"file_purpose": "batch",
"status": {
@ -333,38 +184,82 @@ class CheckBatchCost:
]
},
},
take=MAX_OBJECTS_PER_POLL_CYCLE,
order={"created_at": "asc"},
)
async def _retire_job(self, job: "_ManagedObjectRow", reason: str) -> None:
async def _find_jobs(self, where: Mapping[str, object]) -> Sequence[ManagedObjectRow]:
arrivals: Final = self._poll_arrivals
page: Final = await self._find_job_page(where, arrivals=arrivals)
self._poll_arrivals = not arrivals if page else arrivals
if page:
return page
return await self._find_job_page(where, arrivals=not arrivals)
async def _find_job_page(self, where: Mapping[str, object], *, arrivals: bool) -> Sequence[ManagedObjectRow]:
cursor: Final = self._arrival_cursor if arrivals else self._poll_cursor
if cursor is None:
if arrivals:
self._arrival_before = datetime.now(timezone.utc)
else:
self._poll_before = datetime.now(timezone.utc) - RECENT_BATCH_POLL_WINDOW
before: Final = self._arrival_before if arrivals else self._poll_before
page: Final = await _managed_object_table(self.prisma_client).find_many(
where={
**where,
"created_at": {
"lte": before,
**({"gt": before - RECENT_BATCH_POLL_WINDOW} if arrivals else {}),
},
**(
{
"OR": [
{"created_at": {"gt": cursor[0]}},
{"created_at": cursor[0], "id": {"gt": cursor[1]}},
]
}
if cursor is not None
else {}
),
},
take=MAX_OBJECTS_PER_POLL_CYCLE,
order=[{"created_at": "asc"}, {"id": "asc"}],
)
next_cursor: Final = (page[-1].created_at, page[-1].id) if len(page) == MAX_OBJECTS_PER_POLL_CYCLE else None
if arrivals:
self._arrival_cursor = next_cursor
else:
self._poll_cursor = next_cursor
return page
def _unprocessed_snapshot(self, job: ManagedObjectRow) -> Mapping[str, object]:
return {
"id": job.id,
"updated_at": job.updated_at,
**({"batch_processed": False} if self._has_batch_processed_column else {}),
}
async def _retire_job(self, job: "ManagedObjectRow", reason: str) -> None:
"""
Take a row that can never be costed out of the poll page. Leaving it selectable
would burn one of the MAX_OBJECTS_PER_POLL_CYCLE slots on every future cycle, and
once enough such rows accumulate no newer batch is ever reached. Older schemas
without batch_processed can only be excluded through the status filter.
"""
data: Final = (
{"batch_processed": True}
if self._has_batch_processed_column
else {"status": "stale_expired"}
)
data: Final = {"batch_processed": True} if self._has_batch_processed_column else {"status": "stale_expired"}
try:
await _managed_object_table(self.prisma_client).update(
where={"id": job.id},
retired: Final = await _managed_object_table(self.prisma_client).update_many(
where=self._unprocessed_snapshot(job),
data=data,
)
except Exception as db_err:
verbose_proxy_logger.error(
f"CheckBatchCost: failed to retire uncostable job {job.id} ({reason}): {db_err}"
)
verbose_proxy_logger.error(f"CheckBatchCost: failed to retire uncostable job {job.id} ({reason}): {db_err}")
return
if retired == 0:
return
verbose_proxy_logger.warning(
f"CheckBatchCost: job {job.id} can never be costed ({reason}), "
"so it will no longer be polled"
f"CheckBatchCost: job {job.id} can never be costed ({reason}), so it will no longer be polled"
)
async def _claim_job_for_costing(self, job: "_ManagedObjectRow") -> bool:
async def _claim_job_for_costing(self, job: "ManagedObjectRow", response: "LiteLLMBatch") -> bool:
"""
Atomically flip batch_processed from false to true, returning whether this pod won
the row. Every pod and uvicorn worker schedules its own poller against the shared
@ -378,10 +273,16 @@ class CheckBatchCost:
"""
if not self._has_batch_processed_column:
return True
from litellm_enterprise.proxy.hooks.managed_files import postgres_storable_json
try:
claimed: Final = await _managed_object_table(self.prisma_client).update_many(
where={"id": job.id, "batch_processed": False},
data={"batch_processed": True},
data={
"batch_processed": True,
"status": response.status if response.status != "completed" else "complete",
"file_object": postgres_storable_json(response),
},
)
except Exception as db_err:
verbose_proxy_logger.error(
@ -390,7 +291,9 @@ class CheckBatchCost:
return False
return claimed > 0
async def _release_job_claim(self, job: "_ManagedObjectRow") -> None:
async def _release_job_claim(
self, job: "ManagedObjectRow", original_status: str | None, original_file_object: object
) -> None:
"""Give a claimed row back once billing it failed, so a later poll cycle retries it.
Safe to match on batch_processed=True: while this poller is active the retrieve
@ -402,7 +305,11 @@ class CheckBatchCost:
try:
await _managed_object_table(self.prisma_client).update_many(
where={"id": job.id, "batch_processed": True},
data={"batch_processed": False},
data={
"batch_processed": False,
"status": original_status,
"file_object": json.dumps(original_file_object),
},
)
except Exception as db_err:
verbose_proxy_logger.error(
@ -411,7 +318,7 @@ class CheckBatchCost:
)
@staticmethod
def _has_unified_id_without_model(job: "_ManagedObjectRow") -> bool:
def _has_unified_id_without_model(job: "ManagedObjectRow") -> bool:
"""A unified id that decodes but carries no model_id can never be routed."""
from litellm.proxy.openai_files_endpoints.common_utils import (
convert_b64_uid_to_unified_uid,
@ -436,13 +343,13 @@ class CheckBatchCost:
from litellm.exceptions import NotFoundError
return isinstance(error, (NotFoundError, openai.NotFoundError)) and batch_id in str(error)
return isinstance(error, (NotFoundError, openai.NotFoundError)) and _error_names_resource(error, batch_id)
def _batch_deployment_exists(self, model_id: str) -> bool:
"""A 404 only proves the batch is gone when it came from the batch's own
deployment. Once that deployment leaves the router, default fallbacks can
silently send the retrieve to a provider that never saw the batch, so its
404 must not retire the row; the staleness sweep bounds it instead."""
404 must not retire a potentially recoverable row."""
return self.llm_router.get_deployment(model_id=model_id) is not None
@staticmethod
@ -457,13 +364,13 @@ class CheckBatchCost:
if not output_file_id:
return False
return isinstance(error, (NotFoundError, openai.NotFoundError)) and output_file_id in str(error)
return isinstance(error, (NotFoundError, openai.NotFoundError)) and _error_names_resource(error, output_file_id)
async def _finalize_unbilled_terminal_job(
self, job: "_ManagedObjectRow", response: "LiteLLMBatch"
) -> None:
self, job: "ManagedObjectRow", response: "LiteLLMBatch"
) -> bool:
"""Persist a terminal batch that has nothing billable, converting any raw
provider file ids to managed ids, and take it out of the poll page."""
provider file ids to managed ids, and take it out of the poll page. Returns whether it was persisted."""
try:
from litellm.proxy.openai_files_endpoints.common_utils import (
_is_base64_encoded_unified_file_id,
@ -479,22 +386,24 @@ class CheckBatchCost:
db_batch_object=job,
unified_batch_id=_is_base64_encoded_unified_file_id(job.unified_object_id),
)
update_data: Final[dict] = {
update_data: Final[dict[str, object]] = {
"status": response.status,
"file_object": response.model_dump_json(),
**({"batch_processed": True} if self._has_batch_processed_column else {}),
}
await _managed_object_table(self.prisma_client).update(
where={"id": job.id},
finalized: Final = await _managed_object_table(self.prisma_client).update_many(
where=self._unprocessed_snapshot(job),
data=update_data,
)
verbose_proxy_logger.info(
f"CheckBatchCost: marked job {job.id} as {response.status} in DB"
)
if finalized == 0:
return False
verbose_proxy_logger.info(f"CheckBatchCost: marked job {job.id} as {response.status} in DB")
return True
except Exception as db_err:
verbose_proxy_logger.error(
f"CheckBatchCost: failed to mark job {job.id} as {response.status} in DB: {db_err}"
)
return False
@staticmethod
def _record_error(
@ -505,7 +414,7 @@ class CheckBatchCost:
def _resolve_job_routing(
self,
job: "_ManagedObjectRow",
job: "ManagedObjectRow",
prom_logger: Optional["PrometheusLogger"],
) -> Optional[Tuple[str, str]]:
"""
@ -582,7 +491,7 @@ class CheckBatchCost:
def _resolve_unmanaged_provider_routing(
self,
job: "_ManagedObjectRow",
job: "ManagedObjectRow",
prom_logger: Optional["PrometheusLogger"],
llm_provider: str,
bare_model_name: str,
@ -678,7 +587,7 @@ class CheckBatchCost:
@classmethod
def _get_managed_file_model_name(
cls,
job: "_ManagedObjectRow",
job: "ManagedObjectRow",
deployment_info: "Deployment",
) -> Optional[str]:
"""
@ -698,9 +607,7 @@ class CheckBatchCost:
)
@staticmethod
def _get_input_file_id(job: "_ManagedObjectRow") -> Optional[str]:
import json
def _get_input_file_id(job: "ManagedObjectRow") -> Optional[str]:
from litellm.types.utils import LiteLLMBatch
file_object: Final = job.file_object
@ -719,7 +626,7 @@ class CheckBatchCost:
async def _track_completed_batch_cost(
self,
job: "_ManagedObjectRow",
job: "ManagedObjectRow",
response: "LiteLLMBatch",
model_id: str,
batch_id: str,
@ -735,15 +642,19 @@ class CheckBatchCost:
"""
from litellm.batches.batch_utils import (
calculate_batch_cost_and_usage,
count_error_file_failed_requests,
get_file_content_as_dictionary,
calculate_batch_cost_and_usage,
)
from litellm.files.main import afile_content
from litellm.files.types import FileContentCallOptions
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
from litellm.litellm_core_utils.litellm_logging import Logging as LiteLLMLogging
from litellm.litellm_core_utils.litellm_logging import deployment_pricing_model_info, mask_api_base_credentials
from litellm.litellm_core_utils.litellm_logging import (
deployment_pricing_model_info,
mask_api_base_credentials,
resolve_vertex_location_for_cost,
)
from litellm.proxy.openai_files_endpoints.common_utils import (
_is_base64_encoded_unified_file_id,
)
@ -853,8 +764,7 @@ class CheckBatchCost:
)
except Exception as _e:
verbose_proxy_logger.warning(
f"CheckBatchCost: failed to create managed file ID for "
f"{_file_attr}={_raw_file_id!r}: {_e}"
f"CheckBatchCost: failed to create managed file ID for {_file_attr}={_raw_file_id!r}: {_e}"
)
# Pass the deployment's router-registered pricing (litellm_params custom
@ -873,6 +783,12 @@ class CheckBatchCost:
custom_llm_provider=batch_file_provider,
model_name=model_name,
model_info=deployment_model_info,
vertex_location=resolve_vertex_location_for_cost(
custom_llm_provider=llm_provider,
litellm_params=deployment_info.litellm_params.model_dump(exclude_none=True),
optional_params=None,
model=model_name,
),
)
error_file_failed_requests: Final = await count_error_file_failed_requests(
response,
@ -890,8 +806,9 @@ class CheckBatchCost:
failed_requests=output_file_result.failed_requests + error_file_failed_requests,
)
)
logging_model: Final = batch_result.models[0] if batch_result.models else litellm_model_name or model_id
logging_obj = LiteLLMLogging(
model=batch_result.models[0],
model=logging_model,
messages=[{"role": "user", "content": "<retrieve_batch>"}],
stream=False,
call_type="aretrieve_batch",
@ -911,7 +828,7 @@ class CheckBatchCost:
},
**({"api_base": mask_api_base_credentials(deployment_api_base)} if deployment_api_base else {}),
"metadata": {
**(await self._build_creator_attribution_metadata(job, batch_id)),
**(await self._creator_attribution.batch_metadata(job, batch_id)),
# spend logs read the deployment identity off these metadata keys, so
# without them the batch cost row carries no model_id or model_group
"model_info": {"id": model_id},
@ -922,27 +839,35 @@ class CheckBatchCost:
custom_llm_provider=str(llm_provider) if llm_provider else None,
)
if not await self._claim_job_for_costing(job):
original_status: Final = job.status
original_file_object: Final = job.file_object
if not await self._claim_job_for_costing(job, response.model_copy(update={"usage": batch_result.usage})):
verbose_proxy_logger.info(
f"CheckBatchCost: batch {batch_id} (job {job.id}) was claimed by another pod "
"in this window, so its cost is already being tracked there"
)
return None
try:
await logging_obj.async_success_handler(
result=response,
batch_cost=batch_result.cost,
batch_usage=batch_result.usage,
batch_models=batch_result.models,
batch_successful_requests=batch_result.successful_requests,
batch_failed_requests=batch_result.failed_requests,
batch_prompt_cost=batch_result.prompt_cost,
batch_completion_cost=batch_result.completion_cost,
)
except Exception:
await self._release_job_claim(job)
raise
from litellm.proxy.hooks.proxy_track_cost_callback import track_batch_accounting
with track_batch_accounting(logging_obj.litellm_call_id) as accounting:
try:
await logging_obj.async_success_handler(
result=response,
batch_cost=batch_result.cost,
batch_usage=batch_result.usage,
batch_models=batch_result.models,
batch_successful_requests=batch_result.successful_requests,
batch_failed_requests=batch_result.failed_requests,
batch_prompt_cost=batch_result.prompt_cost,
batch_completion_cost=batch_result.completion_cost,
)
finally:
if accounting.retryable_failure:
await self._release_job_claim(job, original_status, original_file_object)
if accounting.retryable_failure:
return None
# Record batch duration (completed_at - created_at)
if prom_logger and response.completed_at and response.created_at:
@ -956,7 +881,202 @@ class CheckBatchCost:
return model_name, str(llm_provider) if llm_provider else None
async def check_batch_cost(self):
async def _retire_unbilled_terminal_job(
self, job: "ManagedObjectRow", response: "LiteLLMBatch", model_id: str, batch_id: str
) -> None:
"""Persist a terminal batch that has nothing billable, then store its lines, reading its input file by the id
the provider reported rather than the managed id that was persisted."""
provider_input_file_id: Final = response.input_file_id
if await self._finalize_unbilled_terminal_job(job, response):
response.input_file_id = provider_input_file_id
self._store_batch_line_items_in_background(job, response, model_id, batch_id, billed=False)
def _store_batch_line_items_in_background(
self, job: "ManagedObjectRow", response: "LiteLLMBatch", model_id: str, batch_id: str, *, billed: bool
) -> None:
"""Store a retired batch's lines off the poll loop, one batch at a time and only between polls, so neither
pricing the lines nor a slow or hung bucket holds up billing or retiring the other batches. While a poll
cycle's worth of batches is still waiting, the lines of the next one are dropped with an error."""
if not litellm.batch_line_item_callbacks:
return
if len(self._line_item_stores) >= MAX_OBJECTS_PER_POLL_CYCLE:
verbose_proxy_logger.error(
f"CheckBatchCost: {len(self._line_item_stores)} batches are still waiting to store their line items, "
f"so the line items of batch {batch_id} (job {job.id}) were not stored"
)
return
store: Final = asyncio.create_task(
self._store_batch_line_items_in_turn(job, response, model_id, batch_id, billed=billed)
)
self._line_item_stores.add(store)
store.add_done_callback(self._line_item_stores.discard)
async def _store_batch_line_items_in_turn(
self, job: "ManagedObjectRow", response: "LiteLLMBatch", model_id: str, batch_id: str, *, billed: bool
) -> None:
async with self._line_item_store_turn:
await self._poll_idle.wait()
await self._store_batch_line_items(job, response, model_id, batch_id, billed=billed)
async def _store_batch_line_items(
self, job: "ManagedObjectRow", response: "LiteLLMBatch", model_id: str, batch_id: str, *, billed: bool
) -> None:
"""Send every line of a finished batch to batch_line_item_callbacks, one part at a time. A billed batch's
output file is read again here, so nothing waiting for its turn holds a batch's files. Failures, including an
output file that can no longer be read, are logged and never raised, so the batch stays billed and retired
whether or not its lines were stored."""
if not litellm.batch_line_item_callbacks:
return
from litellm.batches.batch_line_items import (
BatchFileProvider,
BatchLineItemSource,
batch_line_item_parts,
dispatch_batch_line_items,
line_item_metadata,
)
from litellm.batches.batch_utils import fetch_batch_managed_file_content
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
from litellm.litellm_core_utils.litellm_logging import (
deployment_pricing_model_info,
resolve_vertex_location_for_cost,
)
try:
deployment_info: Final = self.llm_router.get_deployment(model_id=model_id)
if deployment_info is None:
verbose_proxy_logger.error(
f"CheckBatchCost: deployment {model_id} is no longer in the router, so the line items of batch "
f"{batch_id} (job {job.id}) were not stored"
)
return
litellm_model_name: Final = deployment_info.litellm_params.model
model_name, llm_provider, _, _ = get_llm_provider(
model=litellm_model_name,
custom_llm_provider=deployment_info.litellm_params.custom_llm_provider,
)
cost_provider: Final = cast(BatchFileProvider, llm_provider)
credentials: Final = self.llm_router.get_deployment_credentials_with_provider(model_id) or {}
file_access: Final = {
**credentials,
"_litellm_internal_model_credentials": MappingProxyType(dict(credentials)),
}
output_content: Final = (
await fetch_batch_managed_file_content(
response.output_file_id, custom_llm_provider=cost_provider, litellm_params=file_access
)
if billed and response.output_file_id
else b""
)
error_content: Final = await self._batch_file_content(response.error_file_id, cost_provider, file_access)
if not output_content and not error_content:
return
redact_bodies: Final = await self._creator_turned_off_message_logging(job)
source: Final = BatchLineItemSource(
batch_id=job.unified_object_id,
finished_at=response.completed_at
or response.failed_at
or response.expired_at
or response.cancelled_at
or response.created_at,
model=model_name,
custom_llm_provider=llm_provider,
cost_provider=cost_provider,
model_info=deployment_pricing_model_info(model_id=model_id, deployment_model=litellm_model_name),
metadata=line_item_metadata(await self._creator_attribution.batch_metadata(job, batch_id)),
vertex_location=resolve_vertex_location_for_cost(
custom_llm_provider=llm_provider,
litellm_params=deployment_info.litellm_params.model_dump(exclude_none=True),
optional_params=None,
model=model_name,
),
)
input_content: Final = (
b""
if redact_bodies
else await self._batch_file_content(response.input_file_id, cost_provider, file_access)
)
parts: Final = batch_line_item_parts(
source, input_content, (output_content, error_content), BATCH_LINE_ITEMS_MAX_PART_BYTES
)
handed = 0
# pricing a part's lines is CPU-bound, so each part is built off the event loop
while (payload := await asyncio.to_thread(next, parts, None)) is not None:
await dispatch_batch_line_items(payload, redact_bodies=redact_bodies)
handed += len(payload["lines"])
if handed:
verbose_proxy_logger.info(
f"CheckBatchCost: handed the {handed} line items of batch {batch_id} to batch_line_item_callbacks"
)
except Exception as error: # noqa: BLE001 # storing the lines is best effort, the batch is already billed
verbose_proxy_logger.exception(
f"CheckBatchCost: the line items of batch {batch_id} (job {job.id}) were not stored: {error}"
)
@staticmethod
async def _batch_file_content(
file_id: str | None, custom_llm_provider: "BatchFileProvider", file_access: Mapping[str, object]
) -> bytes:
"""A batch file's raw JSONL, or nothing when the batch has no such file or the provider will not serve it."""
from litellm.batches.batch_utils import fetch_batch_managed_file_content
if not file_id:
return b""
try:
return await fetch_batch_managed_file_content(
file_id, custom_llm_provider=custom_llm_provider, litellm_params=file_access
)
except Exception as error: # noqa: BLE001 # a missing input or error file leaves those lines out, not the rest
verbose_proxy_logger.warning(f"CheckBatchCost: could not read batch file {file_id}: {error}")
return b""
async def _creator_turned_off_message_logging(self, job: "ManagedObjectRow") -> bool:
"""Whether the creating key's or team's logging settings turn off message logging, falling back to the
proxy-wide setting. A batch whose saved credential is not a virtual key's hash, such as one created under JWT
or OAuth2 auth or with the master key, has only its team's and the proxy's settings. A key or team that can no
longer be read counts as turned off."""
from litellm.litellm_core_utils.litellm_logging import is_valid_sha256_hash
from litellm.proxy._types import UserAPIKeyAuth
from litellm.proxy.litellm_pre_call_utils import get_dynamic_logging_metadata
from litellm.proxy.proxy_server import proxy_config
from litellm.secret_managers.main import str_to_bool
api_key: Final = job.api_key if hasattr(job, "api_key") else None
team_id: Final = job.team_id if hasattr(job, "team_id") else None
has_key_row: Final = api_key is not None and is_valid_sha256_hash(api_key)
try:
key_row: Final = (
await _token_table(self.prisma_client).find_unique(where={"token": api_key}) if has_key_row else None
)
team_row: Final = (
await _team_table(self.prisma_client).find_unique(where={"team_id": team_id}) if team_id else None
)
except Exception as error:
verbose_proxy_logger.error(
f"CheckBatchCost: could not read the logging settings of batch job {job.id}'s creator, "
f"so its line items are stored without request and response: {error}"
)
return True
if (has_key_row and key_row is None) or (team_id and team_row is None):
return True
settings: Final = get_dynamic_logging_metadata(
UserAPIKeyAuth(
metadata=dict(row_metadata(key_row)), team_metadata=dict(row_metadata(team_row)), team_id=team_id
),
proxy_config,
)
callback_vars: Final = (settings.callback_vars if settings is not None else None) or {}
creator_setting: Final = str_to_bool(callback_vars.get("turn_off_message_logging"))
return creator_setting if creator_setting is not None else litellm.turn_off_message_logging is True
async def check_batch_cost(self) -> None:
"""Run one poll cycle. Batch line items wait until it has billed and retired every batch it found."""
self._poll_idle.clear()
try:
await self._poll_batches()
finally:
self._poll_idle.set()
async def _poll_batches(self):
"""
Check if the batch JOB has been tracked.
- get all status="validating" and file_purpose="batch" jobs
@ -966,6 +1086,7 @@ class CheckBatchCost:
"""
try:
from litellm.integrations.prometheus import PrometheusLogger
prom_logger = PrometheusLogger.get_instance()
except Exception as e:
verbose_proxy_logger.error(f"CheckBatchCost: could not get Prometheus logger: {e}")
@ -990,7 +1111,7 @@ class CheckBatchCost:
# endpoint may transition a batch to "complete" before
# CheckBatchCost runs. The batch_processed=False filter
# already prevents reprocessing finished batches.
jobs = await _managed_object_table(self.prisma_client).find_many(
jobs = await self._find_jobs(
where={
"file_purpose": "batch",
"batch_processed": False,
@ -1003,8 +1124,6 @@ class CheckBatchCost:
]
},
},
take=MAX_OBJECTS_PER_POLL_CYCLE,
order={"created_at": "asc"},
)
self.batch_processed_support_confirmed = True
except Exception as query_err:
@ -1050,10 +1169,7 @@ class CheckBatchCost:
continue
## RETRIEVE THE BATCH JOB OUTPUT FILE
if (
response.status in PROVIDER_TERMINAL_BATCH_STATUSES
and response.output_file_id is not None
):
if response.status in PROVIDER_TERMINAL_BATCH_STATUSES and response.output_file_id is not None:
try:
tracked = await self._track_completed_batch_cost(
job=job,
@ -1070,11 +1186,11 @@ class CheckBatchCost:
f"CheckBatchCost: output file {response.output_file_id} of batch {batch_id} "
f"does not exist at the provider; retiring job {job.id} unbilled"
)
await self._finalize_unbilled_terminal_job(job, response)
await self._retire_unbilled_terminal_job(job, response, model_id, batch_id)
continue
verbose_proxy_logger.error(
f"CheckBatchCost: failed to track cost for batch {batch_id} "
f"(job {job.id}); leaving it unprocessed so the next poll retries: {tracking_err}"
f"(job {job.id}): {tracking_err}"
)
self._record_error(prom_logger, "cost_tracking_error")
continue
@ -1084,22 +1200,21 @@ class CheckBatchCost:
# Track this job for the final metrics summary
processed_models.append(tracked)
# mark the job as complete
try:
update_data: dict = {
"status": response.status if response.status != "completed" else "complete",
"file_object": response.model_dump_json(),
}
if self._has_batch_processed_column:
update_data["batch_processed"] = True
await _managed_object_table(self.prisma_client).update(
where={"id": job.id},
data=update_data,
)
except Exception as db_err:
verbose_proxy_logger.error(
f"CheckBatchCost: failed to mark job {job.id} complete in DB: {db_err}"
)
if not self._has_batch_processed_column:
try:
await _managed_object_table(self.prisma_client).update(
where={"id": job.id},
data={
"status": response.status if response.status != "completed" else "complete",
"file_object": response.model_dump_json(),
},
)
except Exception as db_err:
verbose_proxy_logger.error(
f"CheckBatchCost: failed to mark job {job.id} complete in DB: {db_err}"
)
continue
self._store_batch_line_items_in_background(job, response, model_id, batch_id, billed=True)
elif response.status in PROVIDER_TERMINAL_BATCH_STATUSES:
from litellm.proxy.openai_files_endpoints.common_utils import (
@ -1112,7 +1227,7 @@ class CheckBatchCost:
f"has not appeared yet; leaving job {job.id} for the next poll cycle"
)
continue
await self._finalize_unbilled_terminal_job(job, response)
await self._retire_unbilled_terminal_job(job, response, model_id, batch_id)
# Record polling run metrics (always, even if nothing was processed)
if prom_logger:

View file

@ -1,12 +1,43 @@
"""
Polls LiteLLM_ManagedObjectTable to check if the response is complete.
Cost tracking is handled by the get-responses call, which prices normally only because the
poll stamps itself with BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN; user-facing reads of the
same route are non-inference and free.
A background response is billed exactly once: whichever client read or poll first sees it
finished moves its row out of a pending status, then logs the body that read already fetched as
one spend event stamped with BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN and the creating key's
attribution, priced against the deployment that served it. The poll's status checks are stamped
with BACKGROUND_RESPONSE_STATUS_POLL_CALL_ORIGIN and client reads are marked managed by the
managed files hook, so both stay free. The spend event carries the id the client holds, the row's
unified_object_id, so its spend log is found under that id with the _background_cost suffix.
The spend event reaches the creating key's or team's own logging callbacks the way that creator's
requests do. Its message content is redacted when the creator's key or team logging turns message
logging off, else as the proxy is set, the way the batch poll decides it. The row is claimed before
the creator is read, so reads that lose the race skip the lookups. A creator whose lookup fails, or
whose logging settings the logger rejects, is still billed, through the proxy's own callbacks only
and with its message content redacted.
A read that finds no row, because the create could not store one, stores it under the reading key
and bills it the same way, so a response the create left untracked is still billed once. A streamed
background create is free on its own; its stream stores the row under the creating key and bills
the finished response through it before the client sees response.completed.
A read the caller retargets with client-side credentials, and a delete of a still-pending response,
settle the bill from a fetch through the configured deployment instead: the client controls what a
retargeted read returns, and a deleted response leaves the poll nothing to fetch.
A row stored before this billing model keeps the encrypted create id as its model_object_id. The
first lookup by that id rekeys it by the provider id, and the poll retires such a row without
billing it once a read has already tracked the response under the provider id.
"""
import uuid
from collections.abc import Awaitable, Callable, Mapping
from dataclasses import dataclass
from datetime import datetime, timedelta, timezone
from typing import TYPE_CHECKING, Dict, Final, Protocol, cast
from types import MappingProxyType
from typing import TYPE_CHECKING, Final, cast
from litellm_enterprise.proxy.common_utils.managed_object_attribution import (
ManagedObjectCreatorAttribution,
ManagedObjectRow,
ResponseCreator,
)
from pydantic import BaseModel, ConfigDict, TypeAdapter
import litellm
from litellm._logging import verbose_proxy_logger
@ -14,36 +45,202 @@ from litellm.constants import (
INTERNAL_CALL_ORIGIN_METADATA_KEY,
MANAGED_OBJECT_STALENESS_CUTOFF_DAYS,
MAX_OBJECTS_PER_POLL_CYCLE,
PENDING_RESPONSE_STATUSES,
STALE_OBJECT_CLEANUP_BATCH_SIZE,
TERMINAL_RESPONSE_STATUSES,
)
from litellm.proxy.hooks.responses_id_security import decrypted_response_id
from litellm.repositories.table_repositories import ManagedObjectRepository
from litellm.responses.utils import ResponsesAPIRequestUtils
from litellm.types.llms.openai import ResponsesAPIResponse
from litellm.types.utils import BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN
from litellm.types.router import Deployment
from litellm.types.utils import (
BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN,
BACKGROUND_RESPONSE_STATUS_POLL_CALL_ORIGIN,
)
if TYPE_CHECKING:
from litellm.litellm_core_utils.litellm_logging import Logging as LiteLLMLogging
from litellm.proxy.proxy_server import ProxyConfig
from litellm.proxy.utils import PrismaClient, ProxyLogging
from litellm.repositories.prisma_protocols import TableActions
from litellm.router import Router
TERMINAL_RESPONSE_STATUSES = frozenset({"completed", "failed", "cancelled", "incomplete"})
_STORED_RESPONSE_ADAPTER: Final[TypeAdapter[dict[str, object]]] = TypeAdapter(dict[str, object])
_CALLBACK_LIST_ADAPTER: Final[TypeAdapter[tuple[object, ...]]] = TypeAdapter(tuple[object, ...])
class _ManagedObjectRow(Protocol):
@property
def id(self) -> str: ...
@property
def unified_object_id(self) -> str: ...
@property
def created_by(self) -> str | None: ...
@property
def file_object(self) -> object: ...
class _RoutedModelInfo(BaseModel):
model_config = ConfigDict(frozen=True)
id: str | None = None
def _managed_object_table(prisma_client: "PrismaClient") -> "TableActions[_ManagedObjectRow]":
class _RoutedDeployment(BaseModel):
model_config = ConfigDict(frozen=True)
model_info: _RoutedModelInfo = _RoutedModelInfo()
_ROUTED_DEPLOYMENTS: Final = TypeAdapter(tuple[_RoutedDeployment, ...])
_RETRIEVE_RESPONSE_CALL_TYPE: Final = "aget_responses"
_CALLBACK_LIST_FIELDS: Final = frozenset({"success_callback", "failure_callback", "litellm_disabled_callbacks"})
_REJECTED_CREATOR_LOGGING: Final[Mapping[str, object]] = MappingProxyType({"turn_off_message_logging": True})
@dataclass(frozen=True, slots=True)
class _PollTarget:
response_id: str
model_metadata: Mapping[str, object]
deployment: Deployment | None
@dataclass(frozen=True, slots=True)
class BilledBackgroundResponse:
response: ResponsesAPIResponse
model: str
custom_llm_provider: str | None
litellm_params: Mapping[str, object]
metadata: Mapping[str, object]
creator_logging: Mapping[str, object]
@dataclass(frozen=True, slots=True)
class BilledCharge:
cost: float | None
logging_obj: "LiteLLMLogging"
BilledResponseLogger = Callable[[BilledBackgroundResponse], Awaitable[BilledCharge]]
UntrackedResponseTracker = Callable[[ResponsesAPIResponse], Awaitable[bool]]
def _without_read_price(response: ResponsesAPIResponse) -> ResponsesAPIResponse:
unpriced: Final = response.model_copy(deep=True)
unpriced.hidden_params = {key: value for key, value in response.hidden_params.items() if key != "response_cost"}
return unpriced
def _string_callbacks(callbacks: object) -> tuple[str, ...]:
if not isinstance(callbacks, list):
return ()
return tuple(
callback for callback in _CALLBACK_LIST_ADAPTER.validate_python(callbacks) if isinstance(callback, str)
)
def _running_proxy_config() -> "ProxyConfig":
from litellm.proxy.proxy_server import proxy_config
return proxy_config
def creator_logging_fields(creator: ResponseCreator, proxy_config: "ProxyConfig | None") -> Mapping[str, object]:
from litellm.proxy._types import UserAPIKeyAuth
from litellm.proxy.litellm_pre_call_utils import key_and_team_logging_fields
if not creator.settings_read:
return _REJECTED_CREATOR_LOGGING
try:
return key_and_team_logging_fields(
user_api_key_dict=UserAPIKeyAuth(
metadata=dict(creator.key_metadata),
team_id=creator.team_id,
team_metadata=dict(creator.team_metadata),
),
proxy_config=proxy_config if proxy_config is not None else _running_proxy_config(),
)
except Exception as e:
verbose_proxy_logger.error(
f"CheckResponsesCost: could not read the creator's logging settings, so only the proxy's "
f"callbacks get the billed event: {e}"
)
return _REJECTED_CREATOR_LOGGING
def _billed_logging_obj(billed: BilledBackgroundResponse, creator_logging: Mapping[str, object]) -> "LiteLLMLogging":
from litellm.litellm_core_utils.litellm_logging import Logging as LiteLLMLogging
from litellm.utils import split_dynamic_success_callbacks
sync_callbacks, async_callbacks = split_dynamic_success_callbacks(
_string_callbacks(creator_logging.get("success_callback"))
)
return LiteLLMLogging(
model=billed.model,
messages=[{"role": "user", "content": "<retrieve_response>"}],
stream=False,
call_type=_RETRIEVE_RESPONSE_CALL_TYPE,
start_time=datetime.now(),
litellm_call_id=str(uuid.uuid4()),
function_id=str(uuid.uuid4()),
dynamic_success_callbacks=list(sync_callbacks),
dynamic_async_success_callbacks=list(async_callbacks) or None,
kwargs={key: value for key, value in creator_logging.items() if key not in _CALLBACK_LIST_FIELDS},
)
def _creator_or_proxy_logging_obj(billed: BilledBackgroundResponse) -> "LiteLLMLogging":
try:
return _billed_logging_obj(billed, billed.creator_logging)
except Exception as e:
verbose_proxy_logger.error(
f"CheckResponsesCost: the creator's logging settings were rejected, so only the proxy's "
f"callbacks get the billed event: {e}"
)
return _billed_logging_obj(billed, _REJECTED_CREATOR_LOGGING)
async def log_billed_background_response(billed: BilledBackgroundResponse) -> BilledCharge:
from litellm.litellm_core_utils.logging_worker import GLOBAL_LOGGING_WORKER
logging_obj: Final = _creator_or_proxy_logging_obj(billed)
logging_obj.update_environment_variables(
litellm_params={**billed.litellm_params, "metadata": dict(billed.metadata)},
optional_params={},
custom_llm_provider=billed.custom_llm_provider,
)
unpriced: Final = _without_read_price(billed.response)
cost: Final = logging_obj.response_cost_calculator(result=unpriced)
end_time: Final = datetime.now()
GLOBAL_LOGGING_WORKER.ensure_initialized_and_enqueue(
async_coroutine=logging_obj.async_success_handler(
result=unpriced, start_time=logging_obj.start_time, end_time=end_time
)
)
logging_obj.handle_sync_success_callbacks_for_async_calls(
result=unpriced, start_time=logging_obj.start_time, end_time=end_time
)
return BilledCharge(cost=cost, logging_obj=logging_obj)
class _DeploymentBaseModel(BaseModel):
model_config = ConfigDict(frozen=True)
base_model: str | None = None
def _deployment_litellm_params(deployment: Deployment) -> Mapping[str, object]:
from litellm.litellm_core_utils.litellm_logging import mask_api_base_credentials
api_base: Final = deployment.litellm_params.api_base
base_model: Final = _DeploymentBaseModel.model_validate(deployment.litellm_params, from_attributes=True).base_model
return MappingProxyType(
{
**({"api_base": mask_api_base_credentials(api_base)} if api_base else {}),
**({"base_model": base_model} if base_model else {}),
}
)
def _deployment_metadata(deployment: Deployment) -> Mapping[str, object]:
return MappingProxyType(
{
"model_info": deployment.model_info.model_dump(exclude_none=True),
"model_group": deployment.model_name,
"deployment": deployment.litellm_params.model,
}
)
def _managed_object_table(prisma_client: "PrismaClient") -> "TableActions[ManagedObjectRow]":
return ManagedObjectRepository(prisma_client).table
@ -51,12 +248,18 @@ def _is_response_gone_at_provider(error: Exception, provider_response_id: str) -
return getattr(error, "status_code", None) == 404 and provider_response_id in str(error)
def _keyed_by_issued_id(job: ManagedObjectRow) -> bool:
return decrypted_response_id(job.model_object_id) != job.model_object_id
class CheckResponsesCost:
def __init__(
self,
proxy_logging_obj: "ProxyLogging",
prisma_client: "PrismaClient",
llm_router: "Router",
log_billed_response: BilledResponseLogger = log_billed_background_response,
proxy_config: "ProxyConfig | None" = None,
):
from litellm.proxy.utils import PrismaClient, ProxyLogging
from litellm.router import Router
@ -64,15 +267,18 @@ class CheckResponsesCost:
self.proxy_logging_obj: ProxyLogging = proxy_logging_obj
self.prisma_client: PrismaClient = prisma_client
self.llm_router: Router = llm_router
self._creator_attribution: Final = ManagedObjectCreatorAttribution(prisma_client, "CheckResponsesCost")
self._log_billed_response: Final = log_billed_response
self._proxy_config: Final = proxy_config
def _resolve_deployment(self, response_id: str) -> bool:
def _resolve_deployment(self, response_id: str) -> Deployment | None:
model_id: str | None = ResponsesAPIRequestUtils.get_model_id_from_response_id(response_id)
return model_id is not None and self.llm_router.get_deployment(model_id=model_id) is not None
return self.llm_router.get_deployment(model_id=model_id) if model_id is not None else None
async def _get_response(
self,
response_id: str,
litellm_metadata: Dict[str, str],
litellm_metadata: dict[str, object],
via_router: bool,
) -> ResponsesAPIResponse:
"""Fetch the upstream response, using deployment credentials when available.
@ -91,9 +297,17 @@ class CheckResponsesCost:
)
return cast(ResponsesAPIResponse, router_response)
async def _expire_stale_rows(
self, cutoff: datetime, batch_size: int
) -> int:
async def _fetch_unbilled(self, target: _PollTarget) -> ResponsesAPIResponse:
return await self._get_response(
response_id=target.response_id,
litellm_metadata={
**target.model_metadata,
INTERNAL_CALL_ORIGIN_METADATA_KEY: BACKGROUND_RESPONSE_STATUS_POLL_CALL_ORIGIN,
},
via_router=target.deployment is not None,
)
async def _expire_stale_rows(self, cutoff: datetime, batch_size: int) -> int:
"""Execute the bounded UPDATE that marks stale rows as 'stale_expired'.
Isolated so it can be swapped / mocked in tests without touching the
@ -140,14 +354,170 @@ class CheckResponsesCost:
f"(older than {MANAGED_OBJECT_STALENESS_CUTOFF_DAYS} days) as stale_expired"
)
def _target(self, response_id: str, model_name: str | None) -> _PollTarget:
return _PollTarget(
response_id=response_id,
model_metadata=MappingProxyType({"model": model_name, "model_group": model_name} if model_name else {}),
deployment=self._resolve_deployment(response_id),
)
def _poll_target(self, job: ManagedObjectRow) -> _PollTarget:
stored_model: Final = _STORED_RESPONSE_ADAPTER.validate_python(job.file_object).get("model")
return self._target(
decrypted_response_id(job.unified_object_id), stored_model if isinstance(stored_model, str) else None
)
async def _claim_for_billing(self, job: ManagedObjectRow) -> bool:
claimed: Final = await _managed_object_table(self.prisma_client).update_many(
# bounded-ok: PENDING_RESPONSE_STATUSES is a two-value constant tuple
where={"id": job.id, "status": {"in": list(PENDING_RESPONSE_STATUSES)}},
data={"status": "completed"},
)
return claimed > 0
async def _release_billing_claim(self, job: ManagedObjectRow) -> None:
await _managed_object_table(self.prisma_client).update_many(
where={"id": job.id, "status": "completed"},
data={"status": job.status or PENDING_RESPONSE_STATUSES[0]},
)
def _billed_response(
self,
job: ManagedObjectRow,
target: _PollTarget,
response: ResponsesAPIResponse,
creator: ResponseCreator,
) -> BilledBackgroundResponse:
deployment: Final = target.deployment
return BilledBackgroundResponse(
response=response.model_copy(update={"id": job.unified_object_id}),
model=deployment.litellm_params.model if deployment is not None else (response.model or ""),
custom_llm_provider=ResponsesAPIRequestUtils.decode_responses_api_response_id(target.response_id).get(
"custom_llm_provider"
),
litellm_params=_deployment_litellm_params(deployment) if deployment is not None else MappingProxyType({}),
metadata=MappingProxyType(
{
**creator.metadata,
INTERNAL_CALL_ORIGIN_METADATA_KEY: BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN,
**target.model_metadata,
**(_deployment_metadata(deployment) if deployment is not None else {}),
**self._requested_model_group(target, job.team_id),
}
),
creator_logging=creator_logging_fields(creator, self._proxy_config),
)
def _requested_model_group(self, target: _PollTarget, team_id: str | None) -> Mapping[str, object]:
requested: Final = target.model_metadata.get("model_group")
if target.deployment is None or not isinstance(requested, str):
return MappingProxyType({})
routed: Final = _ROUTED_DEPLOYMENTS.validate_python(
self.llm_router.get_model_list(model_name=requested, team_id=team_id) or ()
)
serving_id: Final = target.deployment.model_info.id
return MappingProxyType(
{"model_group": requested} if any(row.model_info.id == serving_id for row in routed) else {}
)
async def _bill_once(
self,
job: ManagedObjectRow,
target: _PollTarget,
response: ResponsesAPIResponse,
) -> BilledCharge | None:
if not await self._claim_for_billing(job):
verbose_proxy_logger.info(f"Response {job.unified_object_id} was already billed elsewhere, skipping")
return None
try:
creator: Final = await self._creator_attribution.response_creator(job, job.unified_object_id)
charge: Final = await self._log_billed_response(self._billed_response(job, target, response, creator))
except BaseException:
await self._release_billing_claim(job)
raise
verbose_proxy_logger.info(f"Response {job.unified_object_id} reached a terminal status and was billed")
return charge
async def _job_for_provider_id(self, provider_response_id: str) -> ManagedObjectRow | None:
return await _managed_object_table(self.prisma_client).find_first(
where={"model_object_id": provider_response_id, "file_purpose": "response"}
)
async def _rekey_by_provider_id(self, job: ManagedObjectRow) -> None:
await _managed_object_table(self.prisma_client).update_many(
where={"id": job.id}, data={"model_object_id": decrypted_response_id(job.model_object_id)}
)
async def _tracked_job(self, response_id: str) -> ManagedObjectRow | None:
if tracked := await self._job_for_provider_id(decrypted_response_id(response_id)):
return tracked
stored_by_issued_id: Final = await _managed_object_table(self.prisma_client).find_first(
where={"unified_object_id": response_id, "file_purpose": "response"}
)
if stored_by_issued_id is not None and _keyed_by_issued_id(stored_by_issued_id):
await self._rekey_by_provider_id(stored_by_issued_id)
return stored_by_issued_id
async def _superseded_by_a_provider_id_row(self, job: ManagedObjectRow) -> bool:
if not _keyed_by_issued_id(job):
return False
if await self._job_for_provider_id(decrypted_response_id(job.model_object_id)) is None:
await self._rekey_by_provider_id(job)
return False
await self._claim_for_billing(job)
return True
async def _track_from_deployment(
self, response_id: str, track_untracked: UntrackedResponseTracker | None
) -> BilledCharge | None:
target: Final = self._target(decrypted_response_id(response_id), None)
if track_untracked is None or target.deployment is None:
return None
return await self.bill_if_finished(response_id, await self._fetch_unbilled(target), track_untracked)
async def _start_tracking(
self, response_id: str, response: ResponsesAPIResponse, track_untracked: UntrackedResponseTracker | None
) -> ManagedObjectRow | None:
if track_untracked is None or not await track_untracked(response):
return None
return await self._tracked_job(response_id)
async def bill_if_finished(
self,
response_id: str,
response: ResponsesAPIResponse,
track_untracked: UntrackedResponseTracker | None = None,
) -> BilledCharge | None:
if response.status not in TERMINAL_RESPONSE_STATUSES:
return None
job: Final = await self._tracked_job(response_id) or await self._start_tracking(
response_id, response, track_untracked
)
if job is None or job.status not in PENDING_RESPONSE_STATUSES:
return None
return await self._bill_once(job, self._poll_target(job), response)
async def bill_from_deployment(
self, response_id: str, track_untracked: UntrackedResponseTracker | None = None
) -> BilledCharge | None:
job: Final = await self._tracked_job(response_id)
if job is None:
return await self._track_from_deployment(response_id, track_untracked)
if job.status not in PENDING_RESPONSE_STATUSES:
return None
target: Final = self._poll_target(job)
response: Final = await self._fetch_unbilled(target)
if response.status not in TERMINAL_RESPONSE_STATUSES:
return None
return await self._bill_once(job, target, response)
async def check_responses_cost(self):
"""
Check if background responses are complete and track their cost.
- Get all status="queued" or "in_progress" and file_purpose="response" jobs
- Query the provider to check if response is complete
- Cost is tracked by the get-responses call, billed because the poll is stamped
with BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN
- Mark responses in a terminal state as complete in the database
- Query the provider, unbilled, to check if the response is complete
- Claim each terminal job by moving its row to completed, then log the body that check fetched
as its one billed event, so two pollers racing over the same job bill it once
- Mark responses the provider no longer has (404 through a resolved
deployment) as stale_expired
"""
@ -160,99 +530,53 @@ class CheckResponsesCost:
jobs = await _managed_object_table(self.prisma_client).find_many(
where={
"status": {"in": ["queued", "in_progress"]},
# bounded-ok: PENDING_RESPONSE_STATUSES is a two-value constant tuple
"status": {"in": list(PENDING_RESPONSE_STATUSES)},
"file_purpose": "response",
},
take=MAX_OBJECTS_PER_POLL_CYCLE,
order={"created_at": "asc"},
)
verbose_proxy_logger.debug(f"Found {len(jobs)} response jobs to check")
completed_jobs: Final[list[_ManagedObjectRow]] = []
expired_jobs: Final[list[_ManagedObjectRow]] = []
expired_jobs: Final[list[ManagedObjectRow]] = []
for job in jobs:
unified_object_id = job.unified_object_id
try:
from litellm.proxy.hooks.responses_id_security import (
ResponsesIDSecurity,
)
# Get the stored response object to extract model information
stored_response = job.file_object
model_name = stored_response.get("model", None)
# Decrypt the response ID
responses_id_security, _, _ = ResponsesIDSecurity()._decrypt_response_id(unified_object_id)
# Prepare metadata with model information for cost tracking
litellm_metadata = {
"user_api_key_user_id": job.created_by or "default-user-id",
INTERNAL_CALL_ORIGIN_METADATA_KEY: BACKGROUND_RESPONSE_COST_POLL_CALL_ORIGIN,
}
# Add model information if available
if model_name:
litellm_metadata["model"] = model_name
litellm_metadata["model_group"] = model_name # Use same value for model_group
via_router = self._resolve_deployment(responses_id_security)
if await self._superseded_by_a_provider_id_row(job):
verbose_proxy_logger.info(f"Response {unified_object_id} is billed under its provider id row")
continue
target = self._poll_target(job)
except Exception as e:
verbose_proxy_logger.warning(
f"Skipping job {unified_object_id} due to error: {e}"
)
verbose_proxy_logger.warning(f"Skipping job {unified_object_id} due to error: {e}")
continue
provider_response_id = ResponsesAPIRequestUtils.decode_responses_api_response_id(
responses_id_security
).get("response_id", responses_id_security)
provider_response_id = ResponsesAPIRequestUtils.decode_responses_api_response_id(target.response_id).get(
"response_id", target.response_id
)
try:
response = await self._get_response(
response_id=responses_id_security,
litellm_metadata=litellm_metadata,
via_router=via_router,
)
response = await self._fetch_unbilled(target)
verbose_proxy_logger.debug(
f"Response {unified_object_id} status: {response.status}, model: {model_name}"
)
verbose_proxy_logger.debug(f"Response {unified_object_id} status: {response.status}")
if response.status in TERMINAL_RESPONSE_STATUSES:
await self._bill_once(job, target, response)
except Exception as e:
if via_router and _is_response_gone_at_provider(e, provider_response_id):
if target.deployment is not None and _is_response_gone_at_provider(e, provider_response_id):
verbose_proxy_logger.info(
f"Response {unified_object_id} no longer available at provider (404), marking stale_expired: {e}"
)
expired_jobs.append(job)
continue
verbose_proxy_logger.warning(
f"Skipping job {unified_object_id} due to error: {e}"
)
verbose_proxy_logger.warning(f"Skipping job {unified_object_id} due to error: {e}")
continue
if response.status in TERMINAL_RESPONSE_STATUSES:
verbose_proxy_logger.info(
f"Response {unified_object_id} has terminal status {response.status}, marking as complete"
)
completed_jobs.append(job)
# Mark completed jobs in the database
if len(completed_jobs) > 0:
await _managed_object_table(self.prisma_client).update_many(
# bounded-ok: at most MAX_OBJECTS_PER_POLL_CYCLE rows per cycle, the find_many take above
where={"id": {"in": [job.id for job in completed_jobs]}},
data={"status": "completed"},
)
verbose_proxy_logger.info(
f"Marked {len(completed_jobs)} response jobs as completed"
)
if len(expired_jobs) > 0:
await _managed_object_table(self.prisma_client).update_many(
# bounded-ok: at most MAX_OBJECTS_PER_POLL_CYCLE rows per cycle, the find_many take above
where={"id": {"in": [job.id for job in expired_jobs]}},
data={"status": "stale_expired"},
)
verbose_proxy_logger.info(
f"Marked {len(expired_jobs)} response jobs as stale_expired"
)
verbose_proxy_logger.info(f"Marked {len(expired_jobs)} response jobs as stale_expired")

View file

@ -0,0 +1,381 @@
"""Creator attribution for the spend row a managed-object cost poll writes.
A batch or a background response is billed by a poll job that runs long after the creating
request returned, so the managed object row's creator columns are the only record of which
key, team, organization, user, end user, and tags carry the spend.
"""
from __future__ import annotations
from collections.abc import Mapping
from dataclasses import dataclass
from datetime import datetime
from typing import TYPE_CHECKING, Final, Protocol, TypeGuard, cast
from pydantic import TypeAdapter
from litellm._logging import verbose_proxy_logger
from litellm.constants import CLI_SESSION_KEY_PREFIX
from litellm.repositories.table_repositories import (
DeletedVerificationTokenRepository,
DeprecatedVerificationTokenRepository,
)
from litellm.repositories.team_repository import TeamRepository
from litellm.repositories.user_repository import UserRepository
from litellm.repositories.verification_token_repository import VerificationTokenRepository
if TYPE_CHECKING:
from prisma import models as prisma_models
from litellm.proxy.utils import PrismaClient
from litellm.repositories.prisma_protocols import TableActions
class ManagedObjectRow(Protocol):
@property
def created_at(self) -> datetime: ...
@property
def updated_at(self) -> datetime: ...
@property
def id(self) -> str: ...
@property
def unified_object_id(self) -> str: ...
@property
def model_object_id(self) -> str: ...
@property
def created_by(self) -> str | None: ...
@property
def status(self) -> str | None: ...
@property
def file_object(self) -> object: ...
@property
def org_id(self) -> str | None: ...
@property
def api_key(self) -> str | None: ...
@property
def team_id(self) -> str | None: ...
@property
def request_tags(self) -> object: ...
@property
def end_user_id(self) -> str | None: ...
_TAG_LIST_ADAPTER: Final[TypeAdapter[list[object]]] = TypeAdapter(list[object])
_KEY_OWNER_SCOPE_PREFIX: Final = "key:"
_MAX_ROTATION_HOPS: Final = 16
def _string_tags(request_tags: object) -> list[str] | None:
if not isinstance(request_tags, list):
return None
return [tag for tag in _TAG_LIST_ADAPTER.validate_python(request_tags) if isinstance(tag, str)] or None
def _creator_user_id(created_by: str | None) -> str | None:
if created_by is None or created_by.startswith(_KEY_OWNER_SCOPE_PREFIX):
return None
return created_by
def _user_table(prisma_client: PrismaClient) -> TableActions[prisma_models.LiteLLM_UserTable]:
return UserRepository(prisma_client).table
def _token_table(prisma_client: PrismaClient) -> TableActions[prisma_models.LiteLLM_VerificationToken]:
return VerificationTokenRepository(prisma_client).table
def _team_table(prisma_client: PrismaClient) -> TableActions[prisma_models.LiteLLM_TeamTable]:
return TeamRepository(prisma_client).table
def _rotated_token_table(
prisma_client: PrismaClient,
) -> TableActions[prisma_models.LiteLLM_DeprecatedVerificationToken]:
return DeprecatedVerificationTokenRepository(prisma_client).table
def _deleted_token_table(prisma_client: PrismaClient) -> TableActions[prisma_models.LiteLLM_DeletedVerificationToken]:
return DeletedVerificationTokenRepository(prisma_client).table
def _deleted_team_table(prisma_client: PrismaClient) -> TableActions[prisma_models.LiteLLM_DeletedTeamTable]:
return TeamRepository(prisma_client).deleted_table
@dataclass(frozen=True, slots=True)
class _CreatorRows:
key: prisma_models.LiteLLM_VerificationToken | None
team: prisma_models.LiteLLM_TeamTable | None
user: prisma_models.LiteLLM_UserTable | None
@dataclass(frozen=True, slots=True)
class ResponseCreator:
metadata: Mapping[str, object]
key_metadata: Mapping[str, object]
team_id: str | None
team_metadata: Mapping[str, object]
settings_read: bool = True
class ManagedObjectCreatorAttribution:
"""Rebuilds the spend-tracking metadata of whoever created a managed object, so the poll
job's spend row is attributed like the creating request: key hash, team, organization,
user, aliases, and tags.
Rows created before api_key and request_tags were persisted carry only created_by and
team_id, and fall back to those. A named creating key owns user_api_key_alias; when it
has no alias, or the key has since been rotated or deleted, the field keeps the creating
user's alias, because a resolvable name is more useful on the spend row than a null.
user_api_key_org_id is resolved too: the spend update writer reads it off this metadata
to increment organization spend, so leaving it out drops the cost from org accounting.
A background response also carries its end user, the creating key's project and its alias,
and the key, team, and user budgets and per-model budgets, the way the read that billed it
before did, and never names a key-scoped creator as its user. A key that sets no budget of
its own takes its budget tier's, the same rule auth applies. Its creator's key and team
metadata come along too, so the billed event reaches the key's and team's logging callbacks
like that read did. A rotated key's metadata is read from the live key its rotations lead
to, and a deleted key's or team's from its deleted row. A lookup that fails leaves the
creator's settings unread and the attribution to the rows that still load, so the
response still bills and still counts toward their budgets.
"""
def __init__(self, prisma_client: PrismaClient, poller_name: str) -> None:
self._prisma_client: Final = prisma_client
self._poller_name: Final = poller_name
async def batch_metadata(self, job: ManagedObjectRow, object_id: str) -> Mapping[str, object]:
rows: Final = await self._rows(job, object_id)
return {"user_api_key_user_id": job.created_by, **_shared_metadata(job, rows)}
async def response_creator(self, job: ManagedObjectRow, object_id: str) -> ResponseCreator:
try:
return await self._read_response_creator(job, object_id)
except Exception as e:
verbose_proxy_logger.error(
f"{self._poller_name}: could not read the creator of {object_id}, so it bills without their "
f"settings: {e}"
)
return ResponseCreator(
metadata=_response_creator_metadata(job, await self._rows(job, object_id)),
key_metadata={},
team_id=getattr(job, "team_id", None),
team_metadata={},
settings_read=False,
)
async def _read_response_creator(self, job: ManagedObjectRow, object_id: str) -> ResponseCreator:
api_key: Final = getattr(job, "api_key", None)
team_id: Final = getattr(job, "team_id", None)
rows: Final = _CreatorRows(
key=await self._live_key_row(api_key, job.created_by),
team=await self._live_team_row(team_id),
user=await self._user_row(object_id, _creator_user_id(job.created_by)),
)
return ResponseCreator(
metadata=_response_creator_metadata(job, rows),
key_metadata=row_metadata(rows.key) if rows.key is not None else await self._retired_key_metadata(api_key),
team_id=team_id,
team_metadata=(
row_metadata(rows.team) if rows.team is not None else await self._deleted_team_metadata(team_id)
),
)
async def _retired_key_metadata(self, api_key: str | None) -> Mapping[str, object]:
if not _names_a_key_row(api_key):
return {}
current_token: Final = await self._token_rotated_to(api_key, _MAX_ROTATION_HOPS)
current_key: Final = (
await _token_table(self._prisma_client).find_unique(where={"token": current_token})
if current_token != api_key
else None
)
if current_key is not None:
return row_metadata(current_key)
return row_metadata(
await _deleted_token_table(self._prisma_client).find_first(
where={"token": current_token}, order={"deleted_at": "desc"}
)
)
async def _token_rotated_to(self, token: str, hops_left: int) -> str:
rotation: Final = await _rotated_token_table(self._prisma_client).find_unique(where={"token": token})
if rotation is None or hops_left == 0:
return token
return await self._token_rotated_to(rotation.active_token_id, hops_left - 1)
async def _deleted_team_metadata(self, team_id: str | None) -> Mapping[str, object]:
if not team_id:
return {}
return row_metadata(
await _deleted_team_table(self._prisma_client).find_first(
where={"team_id": team_id}, order={"deleted_at": "desc"}
)
)
async def _rows(self, job: ManagedObjectRow, object_id: str) -> _CreatorRows:
return _CreatorRows(
key=await self._key_row(object_id, getattr(job, "api_key", None), job.created_by),
team=await self._team_row(object_id, getattr(job, "team_id", None)),
user=await self._user_row(object_id, _creator_user_id(job.created_by)),
)
async def _user_row(self, object_id: str, user_id: str | None) -> prisma_models.LiteLLM_UserTable | None:
"""Objects created by a team or service account key carry no user id, and
find_unique(where={"user_id": None}) raises, so those resolve no row."""
if not user_id:
return None
try:
return await _user_table(self._prisma_client).find_unique(where={"user_id": user_id})
except Exception as e:
verbose_proxy_logger.error(f"{self._poller_name}: could not look up user {user_id} for {object_id}: {e}")
return None
async def _key_row(
self, object_id: str, api_key: str | None, created_by: str | None
) -> prisma_models.LiteLLM_VerificationToken | None:
try:
return await self._live_key_row(api_key, created_by)
except Exception as e:
verbose_proxy_logger.error(f"{self._poller_name}: could not look up the creating key for {object_id}: {e}")
return None
async def _live_key_row(
self, api_key: str | None, created_by: str | None
) -> prisma_models.LiteLLM_VerificationToken | None:
if not api_key or _is_cli_session_key(api_key, created_by):
return None
return await _token_table(self._prisma_client).find_unique(
where={"token": api_key}, include={"litellm_budget_table": True, "litellm_project_table": True}
)
async def _team_row(self, object_id: str, team_id: str | None) -> prisma_models.LiteLLM_TeamTable | None:
try:
return await self._live_team_row(team_id)
except Exception as e:
verbose_proxy_logger.error(f"{self._poller_name}: could not look up team {team_id} for {object_id}: {e}")
return None
async def _live_team_row(self, team_id: str | None) -> prisma_models.LiteLLM_TeamTable | None:
if not team_id:
return None
return await _team_table(self._prisma_client).find_unique(where={"team_id": team_id})
def _response_creator_metadata(job: ManagedObjectRow, rows: _CreatorRows) -> Mapping[str, object]:
end_user_id: Final = getattr(job, "end_user_id", None)
return {
"user_api_key_user_id": _creator_user_id(job.created_by),
**_shared_metadata(job, rows),
**({"user_api_key_end_user_id": end_user_id} if end_user_id else {}),
"user_api_key_project_id": getattr(rows.key, "project_id", None),
"user_api_key_project_alias": getattr(getattr(rows.key, "litellm_project_table", None), "project_alias", None),
**_key_budget_fields(rows.key),
**_budget_fields(
rows.team,
"user_api_key_team_spend",
"user_api_key_team_max_budget",
"user_api_key_team_model_max_budget",
),
**_budget_fields(
rows.user,
"user_api_key_user_spend",
"user_api_key_user_max_budget",
"user_api_key_user_model_max_budget",
),
}
def _shared_metadata(job: ManagedObjectRow, rows: _CreatorRows) -> Mapping[str, object]:
api_key: Final = getattr(job, "api_key", None)
team_id: Final = getattr(job, "team_id", None)
key_alias: Final = _key_alias(api_key, job.created_by, rows.key)
team_alias: Final = getattr(rows.team, "team_alias", None)
org_id: Final = getattr(job, "org_id", None) or _organization_id(rows.key) or _organization_id(rows.team)
tags: Final = _string_tags(getattr(job, "request_tags", None))
return {
"user_api_key": api_key,
"user_api_key_hash": api_key,
"user_api_key_team_id": team_id,
**(
{
"user_api_key_user_email": getattr(rows.user, "user_email", None),
"user_api_key_alias": getattr(rows.user, "user_alias", None),
}
if rows.user is not None
else {}
),
**({"user_api_key_alias": key_alias} if key_alias is not None else {}),
**({"user_api_key_team_alias": team_alias} if team_alias is not None else {}),
**({"user_api_key_org_id": org_id} if org_id is not None else {}),
**({"tags": tags} if tags else {}),
}
def row_metadata(row: object) -> Mapping[str, object]:
metadata: Final = getattr(row, "metadata", None)
if not isinstance(metadata, dict):
return {}
return cast(Mapping[str, object], metadata) # cast-ok: prisma Json object column
def _names_a_key_row(api_key: object) -> TypeGuard[str]:
from litellm.litellm_core_utils.litellm_logging import is_valid_sha256_hash
return isinstance(api_key, str) and is_valid_sha256_hash(api_key)
def _is_cli_session_key(api_key: str, created_by: str | None) -> bool:
return bool(created_by) and api_key == f"{CLI_SESSION_KEY_PREFIX}-{created_by}"
def _key_alias(
api_key: str | None, created_by: str | None, key_row: prisma_models.LiteLLM_VerificationToken | None
) -> str | None:
if api_key and _is_cli_session_key(api_key, created_by):
return api_key
return getattr(key_row, "key_alias", None) if key_row is not None else None
def _organization_id(row: object) -> str | None:
return cast("str | None", getattr(row, "organization_id", None)) if row is not None else None
def _budget_fields(row: object, spend_key: str, max_budget_key: str, model_max_budget_key: str) -> Mapping[str, object]:
if row is None:
return {}
return {
spend_key: getattr(row, "spend", None),
max_budget_key: getattr(row, "max_budget", None),
model_max_budget_key: getattr(row, "model_max_budget", None),
}
def _key_budget_fields(key: object) -> Mapping[str, object]:
if key is None:
return {}
tier: Final = getattr(key, "litellm_budget_table", None)
return {
"user_api_key_spend": getattr(key, "spend", None),
"user_api_key_max_budget": _own_or_tier_budget(key, tier, "max_budget"),
"user_api_key_model_max_budget": _own_or_tier_budget(key, tier, "model_max_budget"),
}
def _own_or_tier_budget(key: object, tier: object, field: str) -> object:
own: Final = getattr(key, field, None)
tier_value: Final = getattr(tier, field, None)
return own if tier_value is None or own not in (None, {}) else tier_value

View file

@ -6,6 +6,7 @@ import base64
import json
import time
from collections.abc import Awaitable, Callable, Iterator, Mapping, Sequence
from itertools import chain
from types import MappingProxyType
from typing import (
TYPE_CHECKING,
@ -24,7 +25,9 @@ from uuid import NAMESPACE_URL, uuid5
import httpx
from fastapi import HTTPException
from pydantic import ValidationError
from openai import APIConnectionError, AsyncOpenAI
from openai.types.file_deleted import FileDeleted
from pydantic import BaseModel, JsonValue, TypeAdapter, ValidationError
from typing_extensions import ReadOnly, Unpack
import litellm
@ -35,17 +38,18 @@ from litellm.caching.caching import DualCache
from litellm.constants import (
BATCH_OUTPUT_FILE_FALLBACK_RETRY_AFTER_SECONDS,
BATCH_OUTPUT_FILE_LOOKUP_TIMEOUT_SECONDS,
MANAGED_BACKGROUND_RESPONSE_METADATA_KEY,
MAX_FILE_LIST_LIMIT,
PENDING_RESPONSE_STATUSES,
)
from litellm.files.types import FileRetrieveCallOptions, FileRetrieveProvider
from litellm.integrations.custom_logger import CustomLogger
from litellm.litellm_core_utils.core_helpers import require_attributes
from litellm.litellm_core_utils.hidden_params import HIDDEN_PARAMS_ATTR
from litellm.litellm_core_utils.internal_call_metadata import requests_background
from litellm.litellm_core_utils.prompt_templates.common_utils import (
extract_file_metadata,
)
from openai import APIConnectionError, AsyncOpenAI
from openai.types.file_deleted import FileDeleted
from litellm.llms.base_llm.files.storage_backend_factory import get_storage_backend
from litellm.llms.base_llm.files.transformation import BaseFileEndpoints
from litellm.llms.base_llm.managed_resources.isolation import (
@ -54,6 +58,7 @@ from litellm.llms.base_llm.managed_resources.isolation import (
can_access_resource,
resolve_resource_owner_id,
)
from litellm.llms.xai.batches.transformation import is_xai_batch_results_id
from litellm.proxy._types import (
CallTypes,
LiteLLM_ManagedFileTable,
@ -87,6 +92,7 @@ from litellm.proxy.pass_through_endpoints.llm_provider_handlers.batch_attributio
request_tags_from_metadata,
)
from litellm.repositories.managed_file_repository import ManagedFileRepository
from litellm.responses.utils import ResponsesAPIRequestUtils
from litellm.types.llms.openai import ( # pyright: ignore[reportAttributeAccessIssue]
AllMessageValues,
AsyncCursorPage,
@ -99,9 +105,11 @@ from litellm.types.llms.openai import ( # pyright: ignore[reportAttributeAccess
ResponsesAPIResponse,
)
from litellm.types.utils import (
ADDRESSED_RESPONSE_ID_FIELD,
CallTypesLiteral,
LiteLLMBatch,
LiteLLMFineTuningJob,
LlmProviders,
LLMResponseTypes,
SpecialEnums,
)
@ -126,7 +134,7 @@ if TYPE_CHECKING:
from litellm.proxy.utils import InternalUsageCache as _InternalUsageCache
from litellm.proxy.utils import PrismaClient as _PrismaClient
Span = Union[_Span, Any]
Span = _Span
InternalUsageCache = _InternalUsageCache
PrismaClient = _PrismaClient
else:
@ -194,6 +202,32 @@ def _is_transient_file_retrieve_error(error: Exception) -> bool:
return isinstance(error, (httpx.TransportError, APIConnectionError, asyncio.TimeoutError))
_MEASURED_SIZE_SAVE_ATTEMPTS: Final = 3
def _stored_json_text(raw_file_object: object) -> str:
"""JSON text of a stored file object exactly as read, for a write that must not race a newer one."""
if isinstance(raw_file_object, BaseModel):
return raw_file_object.model_dump_json()
return json.dumps(raw_file_object)
def _with_measured_size(file_object: OpenAIFileObject, size_bytes: int | None) -> OpenAIFileObject:
"""Fill a file object whose provider reported no size (0 bytes) with a size measured elsewhere."""
if not size_bytes or file_object.bytes:
return file_object
return file_object.model_copy(update={"bytes": size_bytes})
def _over_stored_object(file_object: OpenAIFileObject, stored_object: OpenAIFileObject | None) -> OpenAIFileObject:
"""Refreshed details for a stored row, keeping its size when none is reported and its stored created_at."""
if stored_object is None:
return file_object
return _with_measured_size(file_object, stored_object.bytes).model_copy(
update={"created_at": stored_object.created_at}
)
def _proxy_llm_router() -> Router | None:
import litellm.proxy.proxy_server as proxy_server_module
@ -252,6 +286,23 @@ def _has_provider_file_retrieve_route(
return bool(router_credentials) or _model_name_file_retrieve_provider(model_name) is not None
def _is_xai_batch_output(
*,
llm_router: Router | None,
model_id: str | None,
model_name: str | None,
provider_file_id: str,
) -> bool:
"""Whether a batch output row holds an xAI batch's results, from its batch id and its deployment's provider."""
if not is_xai_batch_results_id(provider_file_id):
return False
router_credentials: Final = _provider_file_retrieve_credentials(llm_router=llm_router, model_id=model_id)
provider: Final = (router_credentials or {}).get("custom_llm_provider") or _model_name_file_retrieve_provider(
model_name
)
return provider == LlmProviders.XAI.value
class _ManagedFileRow(Protocol):
unified_file_id: str
file_object: OpenAIFileObject
@ -301,6 +352,8 @@ class _ManagedObjectTableActions(Protocol):
async def update_many(self, where: Mapping[str, object], data: Mapping[str, object]) -> int: ...
async def create(self, data: Mapping[str, object]) -> "PrismaManagedObjectRow": ...
class _ManagedResourceDatabase(Protocol):
@property
@ -347,6 +400,39 @@ def _managed_object_table(prisma_client: _ManagedResourcePrismaClient) -> _Manag
return prisma_client.db.litellm_managedobjecttable
_JSON_VALUE_ADAPTER: Final[TypeAdapter[JsonValue]] = TypeAdapter(JsonValue)
_POSTGRES_UNSTORABLE_CHARACTER: Final = "\x00"
def _storable_text(text: str) -> str:
return text.replace(_POSTGRES_UNSTORABLE_CHARACTER, "")
def _storable_json_value(value: JsonValue) -> JsonValue:
if isinstance(value, str):
return _storable_text(value)
if isinstance(value, list):
return [_storable_json_value(item) for item in value]
if isinstance(value, dict):
return {_storable_text(key): _storable_json_value(item) for key, item in value.items()}
return value
def postgres_storable_json(file_object: BaseModel) -> str:
dumped: Final = file_object.model_dump_json()
if "\\u0000" not in dumped:
return dumped
return _JSON_VALUE_ADAPTER.dump_json(_storable_json_value(_JSON_VALUE_ADAPTER.validate_json(dumped))).decode()
def _without_model_output(
file_object: LiteLLMBatch | LiteLLMFineTuningJob | ResponsesAPIResponse,
) -> LiteLLMBatch | LiteLLMFineTuningJob | ResponsesAPIResponse:
return (
file_object.model_copy(update={"output": []}) if isinstance(file_object, ResponsesAPIResponse) else file_object
)
def _storage_metadata_of(file_object: OpenAIFileObject | None) -> Mapping[str, str]:
hidden_params: Final = cast( # cast-ok: _hidden_params is an untyped attribute the upload path sets
"Mapping[str, object]", getattr(file_object, "_hidden_params", None) or {}
@ -360,10 +446,59 @@ def _storage_metadata_of(file_object: OpenAIFileObject | None) -> Mapping[str, s
)
_MANAGED_RESPONSE_READ_CALL_TYPES: Final = frozenset({"aget_responses", "acancel_responses"})
_REQUEST_FIELDS: Final = TypeAdapter(dict[str, object])
def _addresses_a_proxy_issued_response(data: Mapping[str, object]) -> bool:
from litellm.proxy.hooks.responses_id_security import decrypted_response_id
addressed_id: Final = data.get(ADDRESSED_RESPONSE_ID_FIELD) or data.get("response_id")
return (
isinstance(addressed_id, str)
and ResponsesAPIRequestUtils.get_model_id_from_response_id(decrypted_response_id(addressed_id)) is not None
)
def _is_streamed_background_create(request: Mapping[str, object]) -> bool:
return (
request.get("stream") is True
and isinstance(request.get("litellm_metadata"), dict)
and requests_background(request)
)
def _with_managed_background_marker(request: Mapping[str, object]) -> dict[str, object]:
metadata_field: Final = "litellm_metadata" if isinstance(request.get("litellm_metadata"), dict) else "metadata"
existing_metadata: Final = request.get(metadata_field)
return {
**request,
metadata_field: {
**(_REQUEST_FIELDS.validate_python(existing_metadata) if isinstance(existing_metadata, dict) else {}),
MANAGED_BACKGROUND_RESPONSE_METADATA_KEY: True,
},
}
_MANAGED_FILES_TARGET: Final = "managed_files"
_PROVIDER_FILE_ID_KWARGS: Final = MappingProxyType(
{
CallTypes.acreate_batch: ("input_file_id",),
CallTypes.acreate_fine_tuning_job: ("training_file", "validation_file"),
}
)
_PROVIDER_FILE_ID_KEYS: Final = tuple(dict.fromkeys(chain.from_iterable(_PROVIDER_FILE_ID_KWARGS.values())))
_PROVIDER_FILE_RETRIEVE_RETRY_DELAYS_SECONDS: Final = (0.5, 1.0, 2.0)
class _SupportsKeyedGet(Protocol):
def get(self, key: str, default: object, /) -> object: ...
def _keyed_get(source: object, key: str, default: object = None) -> object:
return require_attributes(source, _SupportsKeyedGet, "get").get(key, default)
class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
# Class variables or attributes
def __init__(
@ -484,11 +619,14 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
) -> None:
"""Persist a managed object row, caching it and upserting it in the DB.
persist_attribution is set only by the batch create, which is the one caller
that can speak for the creator; it gates the api_key and request_tags columns
that CheckBatchCost bills against, so a later poll or retrieve of the same
batch cannot record itself as the paying key. Like created_by and team_id,
both are written only in the upsert create branch, never on update.
persist_attribution is set only by the batch and background response creates,
the callers that can speak for the creator; it gates the api_key, request_tags, and,
for a response, end_user_id columns that the cost poll jobs bill against, so a later poll or
retrieve of the same object cannot record itself as the paying key. Like created_by and team_id,
they are written only in the upsert create branch, never on update.
A stored response keeps no model output: billing reads only its status and model from
the row, so the row never holds generated content.
create_if_missing is cleared by callers that observe a batch they did not
create, such as a poll. They still refresh status and file_object, but a
@ -501,11 +639,12 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
It is written only in the upsert create branch.
"""
verbose_logger.info(f"Storing LiteLLM Managed {file_purpose} object with id={unified_object_id} in cache")
stored_object: Final = _without_model_output(file_object)
litellm_managed_object = LiteLLM_ManagedObjectTable(
unified_object_id=unified_object_id,
model_object_id=model_object_id,
file_purpose=file_purpose,
file_object=file_object,
file_object=stored_object,
)
await self.internal_usage_cache.async_set_cache(
key=unified_object_id,
@ -516,32 +655,35 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
from prisma import Json
api_key = LiteLLMProxyRequestSetup.get_logged_api_key(user_api_key_dict) or None
end_user_id: Final = user_api_key_dict.end_user_id if file_purpose == "response" else None
attribution_columns = (
{
**({"api_key": api_key} if api_key is not None else {}),
**({"request_tags": Json(list(request_tags))} if request_tags else {}),
**({"request_tags": Json([_storable_text(tag) for tag in request_tags])} if request_tags else {}),
**({"end_user_id": end_user_id} if end_user_id else {}),
}
if persist_attribution
else {}
)
# FIX: Update status and file_object on every operation to keep state in sync
update_columns: Final = {
"file_object": file_object.model_dump_json(),
"file_object": postgres_storable_json(stored_object),
"status": file_object.status,
"updated_by": user_api_key_dict.user_id,
}
managed_object_table: Final = _managed_object_table(self.prisma_client)
if not create_if_missing:
await _managed_object_table(self.prisma_client).update_many(
await managed_object_table.update_many(
where={"unified_object_id": unified_object_id},
data=update_columns,
)
return
await _managed_object_table(self.prisma_client).upsert(
await managed_object_table.upsert(
where={"unified_object_id": unified_object_id},
data={
"create": {
"unified_object_id": unified_object_id,
"file_object": file_object.model_dump_json(),
"file_object": postgres_storable_json(stored_object),
"model_object_id": model_object_id,
"file_purpose": file_purpose,
"created_by": resolve_resource_owner_id(user_api_key_dict),
@ -556,6 +698,42 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
},
)
async def track_untracked_background_response(
self,
response_id: str,
response: ResponsesAPIResponse,
user_api_key_dict: UserAPIKeyAuth,
request_tags: Sequence[str] | None = None,
) -> bool:
if not _addresses_a_proxy_issued_response({"response_id": response_id}):
return False
from prisma import Json
from prisma.errors import UniqueViolationError
from litellm.proxy.hooks.responses_id_security import decrypted_response_id
api_key: Final = LiteLLMProxyRequestSetup.get_logged_api_key(user_api_key_dict) or None
try:
await _managed_object_table(self.prisma_client).create(
data={
"unified_object_id": response_id,
"model_object_id": decrypted_response_id(response_id),
"file_purpose": "response",
"file_object": postgres_storable_json(_without_model_output(response)),
"status": PENDING_RESPONSE_STATUSES[0],
"created_by": resolve_resource_owner_id(user_api_key_dict),
"team_id": user_api_key_dict.team_id,
"org_id": await self._resolve_creator_org_id(user_api_key_dict),
"updated_by": user_api_key_dict.user_id,
**({"api_key": api_key} if api_key is not None else {}),
**({"end_user_id": user_api_key_dict.end_user_id} if user_api_key_dict.end_user_id else {}),
**({"request_tags": Json([_storable_text(tag) for tag in request_tags])} if request_tags else {}),
}
)
except UniqueViolationError:
return False
return True
@with_service_target(_MANAGED_FILES_TARGET)
async def get_unified_file_id(
self, file_id: str, litellm_parent_otel_span: Optional[Span] = None
@ -845,6 +1023,57 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
litellm_parent_otel_span=None,
)
async def _save_xai_refreshed_file_object(
self,
stored: LiteLLM_ManagedFileTable,
file_object: OpenAIFileObject,
) -> OpenAIFileObject | None:
"""Save an xAI batch output's refreshed details, keeping a known size, only over the row as read; drop the cached copy if a newer write won."""
stored_object: Final = _parse_managed_file_object(stored.file_object, stored.unified_file_id)
sized_file_object: Final = _over_stored_object(file_object, stored_object)
if not await ManagedFileRepository(self.prisma_client).update_file_object(
stored.unified_file_id,
sized_file_object,
if_file_object=_stored_json_text(stored.file_object) if stored.file_object is not None else None,
):
await self.internal_usage_cache.async_set_cache(
key=stored.unified_file_id, value=None, litellm_parent_otel_span=None
)
return None
await self.internal_usage_cache.async_set_cache(
key=stored.unified_file_id,
value=stored.model_copy(update={"file_object": sized_file_object}).model_dump(),
litellm_parent_otel_span=None,
)
return sized_file_object
async def _save_batch_output_file_object(
self,
stored: LiteLLM_ManagedFileTable,
file_object: OpenAIFileObject,
size_bytes: int | None,
) -> None:
"""Save an xAI batch output's details; after losing to a refresh, size the refreshed details instead of replacing them."""
current_row: LiteLLM_ManagedFileTable | None = stored # rebind-ok: the row as last read
details: OpenAIFileObject = file_object # rebind-ok: a concurrent refresh's details replace the job's copy
for _ in range(_MEASURED_SIZE_SAVE_ATTEMPTS):
if current_row is None:
return
saved = await self._save_xai_refreshed_file_object( # rebind-ok: per attempt
current_row, _with_measured_size(details, size_bytes)
)
if saved is not None or not size_bytes:
return
latest_row = await ManagedFileRepository( # rebind-ok: per attempt
self.prisma_client, use_writer=True
).table.find_first(where={"unified_file_id": stored.unified_file_id})
current_row = (
None if latest_row is None else LiteLLM_ManagedFileTable.model_validate(latest_row.model_dump())
)
latest_object = None if current_row is None else current_row.file_object # rebind-ok: per attempt
if latest_object is not None and not latest_object.litellm_details_fallback:
details = latest_object
async def store_batch_output_file(
self,
*,
@ -885,6 +1114,18 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
)
return
if stored_object is not None and not stored_object.litellm_details_fallback:
if (
stored_file is not None
and size_bytes
and not stored_object.bytes
and _is_xai_batch_output(
llm_router=_proxy_llm_router(),
model_id=model_id,
model_name=model_name,
provider_file_id=provider_file_id,
)
):
await self._save_batch_output_file_object(stored_file, stored_object, size_bytes)
return
fallback_written_recently: Final = (
@ -924,13 +1165,22 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
)
)
xai_batch_output: Final = _is_xai_batch_output(
llm_router=_proxy_llm_router(),
model_id=model_id,
model_name=model_name,
provider_file_id=provider_file_id,
)
if stored_file is not None and xai_batch_output:
await self._save_batch_output_file_object(stored_file, file_object, size_bytes)
return
if stored_file is not None:
await self._save_refreshed_file_object(stored_file, file_object)
return
await self.store_unified_file_id(
file_id=unified_file_id,
file_object=file_object,
file_object=_with_measured_size(file_object, size_bytes) if xai_batch_output else file_object,
litellm_parent_otel_span=litellm_parent_otel_span,
model_mappings={model_id: provider_file_id} if model_id else {},
user_api_key_dict=owner,
@ -1258,6 +1508,9 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
unified_vs_ids = self.get_vector_store_ids_from_file_search_tools(tools)
if unified_vs_ids:
await self.check_vector_store_ids_access(unified_vs_ids, user_api_key_dict)
create_request: Final = _REQUEST_FIELDS.validate_python(data)
if _is_streamed_background_create(create_request):
return _with_managed_background_marker(create_request)
elif call_type == CallTypes.afile_content.value:
retrieve_file_id = cast(Optional[str], data.get("file_id"))
potential_file_id = _is_base64_encoded_unified_file_id(retrieve_file_id) if retrieve_file_id else False
@ -1315,6 +1568,10 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
data[accessor_key] = get_batch_id_from_unified_batch_id(potential_llm_object_id)
elif retrieve_object_id and accessor_key == "batch_id":
await self.enforce_batch_object_access(retrieve_object_id, user_api_key_dict)
elif call_type in _MANAGED_RESPONSE_READ_CALL_TYPES:
request: Final = _REQUEST_FIELDS.validate_python(data)
if _addresses_a_proxy_issued_response(request):
return _with_managed_background_marker(request)
elif call_type == CallTypes.acreate_fine_tuning_job.value:
fine_tuning_file_ids: Final = tuple(
file_id
@ -1322,6 +1579,10 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
if isinstance(file_id, str) and file_id
)
await self.check_file_ids_access(fine_tuning_file_ids, user_api_key_dict)
if fine_tuning_file_ids:
data["model_file_id_mapping"] = await self.get_model_file_id_mapping(
fine_tuning_file_ids, user_api_key_dict.parent_otel_span
)
return data
@ -1336,50 +1597,49 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
if request_kwargs is None:
return healthy_deployments
input_file_id = cast(Optional[str], request_kwargs.get("input_file_id"))
model_file_id_mapping = cast(
Optional[Dict[str, Dict[str, str]]],
request_kwargs.get("model_file_id_mapping"),
request: Final = cast(Mapping[str, object], request_kwargs)
model_file_id_mapping: Final = cast(Mapping[str, Mapping[str, str]], request.get("model_file_id_mapping") or {})
requested_file_ids: Final = (request.get(key) for key in _PROVIDER_FILE_ID_KEYS)
model_ids_per_file: Final = tuple(
frozenset(model_file_id_mapping[file_id])
for file_id in requested_file_ids
if isinstance(file_id, str) and file_id in model_file_id_mapping
)
allowed_model_ids: Final[frozenset[str]] = (
model_ids_per_file[0].intersection(*model_ids_per_file[1:]) if model_ids_per_file else frozenset()
)
allowed_model_ids = []
if input_file_id and model_file_id_mapping:
model_id_dict = model_file_id_mapping.get(input_file_id, {})
allowed_model_ids = list(model_id_dict.keys())
if len(allowed_model_ids) == 0:
return healthy_deployments
return [
holders: Final[list[dict[str, object]]] = [
deployment
for deployment in healthy_deployments
if deployment.get("model_info", {}).get("id") in allowed_model_ids
]
return holders if holders or "input_file_id" in request else healthy_deployments
async def async_pre_call_deployment_hook(
self, kwargs: Dict[str, Any], call_type: Optional[CallTypes]
self, kwargs: dict[str, object], call_type: Optional[CallTypes]
) -> Optional[dict]:
"""
Allow modifying the request just before it's sent to the deployment.
"""
accessor_key: Optional[str] = None
if call_type and call_type == CallTypes.acreate_batch:
accessor_key = "input_file_id"
elif call_type and call_type == CallTypes.acreate_fine_tuning_job:
accessor_key = "training_file"
else:
accessor_keys: Final = _PROVIDER_FILE_ID_KWARGS.get(call_type) if call_type else None
if accessor_keys is None:
return kwargs
if accessor_key:
model_file_id_mapping = cast(Optional[Dict[str, Dict[str, str]]], kwargs.get("model_file_id_mapping"))
# model_info may be at top-level or nested under litellm_metadata
# (batch/file operations use litellm_metadata)
model_id = cast(Optional[str], _keyed_get(kwargs.get("model_info", {}), "id"))
if model_id is None:
model_id = cast(
Optional[str],
_keyed_get(_keyed_get(kwargs.get("litellm_metadata", {}), "model_info", {}), "id"),
)
for accessor_key in accessor_keys:
input_file_id = cast(Optional[str], kwargs.get(accessor_key))
model_file_id_mapping = cast(Optional[Dict[str, Dict[str, str]]], kwargs.get("model_file_id_mapping"))
# model_info may be at top-level or nested under litellm_metadata
# (batch/file operations use litellm_metadata)
model_id = cast(Optional[str], kwargs.get("model_info", {}).get("id", None))
if model_id is None:
model_id = cast(
Optional[str],
kwargs.get("litellm_metadata", {}).get("model_info", {}).get("id", None),
)
mapped_file_id: Optional[str] = None
if input_file_id and model_file_id_mapping and model_id:
mapped_file_id = model_file_id_mapping.get(input_file_id, {}).get(model_id, None)
@ -1573,7 +1833,7 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
),
)
async def get_model_file_id_mapping(self, file_ids: List[str], litellm_parent_otel_span: Span) -> dict:
async def get_model_file_id_mapping(self, file_ids: Sequence[str], litellm_parent_otel_span: Span | None) -> dict:
"""
Get model-specific file IDs for a list of proxy file IDs.
Returns a dictionary mapping litellm_proxy/ file_id -> model_id -> model_file_id
@ -1615,7 +1875,7 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
llm_router: Optional[Router],
_create_file_request: CreateFileRequest,
target_model_names_list: List[str],
litellm_parent_otel_span: Span,
litellm_parent_otel_span: Span | None,
) -> List[OpenAIFileObject]:
if llm_router is None:
raise Exception("LLM Router not initialized. Ensure models added to proxy.")
@ -1631,7 +1891,7 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
create_file_request: CreateFileRequest,
llm_router: Router,
target_model_names_list: List[str],
litellm_parent_otel_span: Span,
litellm_parent_otel_span: Span | None,
user_api_key_dict: UserAPIKeyAuth,
) -> OpenAIFileObject:
responses = await self.create_file_for_each_model(
@ -1698,7 +1958,7 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
file_objects: List[OpenAIFileObject],
create_file_request: CreateFileRequest,
internal_usage_cache: InternalUsageCache,
litellm_parent_otel_span: Span,
litellm_parent_otel_span: Span | None,
target_model_names_list: List[str],
) -> OpenAIFileObject:
## GET THE FILE TYPE FROM THE CREATE FILE REQUEST
@ -1943,8 +2203,20 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
)
if refreshed_file_object is None:
return _public_file_object(file_object, file_id)
await self._save_refreshed_file_object(stored_file_object, refreshed_file_object)
return _public_file_object(refreshed_file_object, file_id)
if not _is_xai_batch_output(
llm_router=llm_router if llm_router is not None else _proxy_llm_router(),
model_id=model_id,
model_name=model_id,
provider_file_id=provider_file_id,
):
await self._save_refreshed_file_object(stored_file_object, refreshed_file_object)
return _public_file_object(refreshed_file_object, file_id)
saved_file_object: Final = await self._save_xai_refreshed_file_object(
stored_file_object, refreshed_file_object
)
return _public_file_object(
saved_file_object or _over_stored_object(refreshed_file_object, file_object), file_id
)
except Exception as error:
verbose_logger.warning(
"Failed to refresh batch file object for "
@ -2389,7 +2661,7 @@ class _PROXY_LiteLLMManagedFiles(CustomLogger, BaseFileEndpoints):
# Continue with other files even if one fails
continue
def _get_content_type_from_file_object(self, file_object: Optional[Any]) -> str:
def _get_content_type_from_file_object(self, file_object: object) -> str:
"""
Determine content type from file object.

View file

@ -29,7 +29,7 @@ if TYPE_CHECKING:
from litellm.proxy.utils import InternalUsageCache as _InternalUsageCache
from litellm.proxy.utils import PrismaClient as _PrismaClient
Span = Union[_Span, Any]
Span = _Span
InternalUsageCache = _InternalUsageCache
PrismaClient = _PrismaClient
else:
@ -107,7 +107,7 @@ class _PROXY_LiteLLMManagedVectorStores(
llm_router: Router,
model: str,
request_data: dict[str, object] | VectorStoreCreateOptionalRequestParams,
litellm_parent_otel_span: Span,
litellm_parent_otel_span: Span | None,
) -> VectorStoreCreateResponse:
"""
Create a vector store for a specific model.
@ -134,7 +134,7 @@ class _PROXY_LiteLLMManagedVectorStores(
create_request: VectorStoreCreateOptionalRequestParams,
llm_router: Router,
target_model_names_list: List[str],
litellm_parent_otel_span: Span,
litellm_parent_otel_span: Span | None,
user_api_key_dict: UserAPIKeyAuth,
) -> VectorStoreCreateResponse:
"""

View file

@ -18,8 +18,8 @@ from fastapi.responses import HTMLResponse, RedirectResponse, Response
from pydantic import ConfigDict, Field, SecretStr, TypeAdapter, ValidationError
from litellm.llms.custom_httpx.http_handler import get_async_httpx_client
from litellm.proxy._experimental.mcp_server.oauth_utils import get_request_base_url
from litellm.proxy._types import LiteLLM_UserTable, LitellmUserRoles, UserAPIKeyAuth
from litellm.proxy.mcp_server.oauth_utils import get_request_base_url
from litellm.types.llms.base import LiteLLMBaseModel
from litellm.types.proxy.auth.auth_checks import UserNotFoundError
@ -122,7 +122,7 @@ def _cookie_name(token: str) -> str:
async def _session_user(request: Request) -> str | None:
from litellm.proxy._experimental.mcp_server.byok_oauth_endpoints import (
from litellm.proxy.mcp_server.byok_oauth_endpoints import (
get_authenticated_browser_user_id,
)

View file

@ -36,6 +36,7 @@ from litellm.repositories.budget_repository import BudgetRepository
from litellm.repositories.object_permission_repository import ObjectPermissionRepository
from litellm.repositories.prisma_protocols import TableActions
from litellm.repositories.project_repository import ProjectRepository
from litellm.repositories.table_repositories import PrismaTableRepository
from litellm.repositories.team_repository import TeamRepository
from litellm.repositories.user_repository import UserRepository
from litellm.repositories.verification_token_repository import VerificationTokenRepository
@ -73,6 +74,16 @@ def _verification_token_table(
return VerificationTokenRepository(prisma_client).table
class _WriterVerificationTokenRepository(PrismaTableRepository["prisma_models.LiteLLM_VerificationToken"]):
table_name = "litellm_verificationtoken"
def _writer_verification_token_table(
prisma_client: PrismaClient,
) -> TableActions["prisma_models.LiteLLM_VerificationToken"]:
return _WriterVerificationTokenRepository(prisma_client, use_writer=True).table
def _budget_table(prisma_client: PrismaClient) -> TableActions["prisma_models.LiteLLM_BudgetTable"]:
return BudgetRepository(prisma_client).table
@ -849,7 +860,7 @@ async def update_project(
else None
)
if current_project is not None and data.team_id != current_project.team_id:
mismatched_key_count: Final = await prisma_client.writer_db.litellm_verificationtoken.count(
mismatched_key_count: Final = await _writer_verification_token_table(prisma_client).count(
where={
"project_id": data.project_id,
"OR": [{"team_id": {"not": data.team_id}}, {"team_id": None}],

View file

@ -14,6 +14,7 @@ from collections.abc import Mapping
from typing import TYPE_CHECKING, Final, List, Optional, Protocol
from fastapi import APIRouter, Depends, HTTPException
from pydantic import TypeAdapter
import litellm
from litellm._logging import verbose_proxy_logger
@ -37,6 +38,7 @@ if TYPE_CHECKING:
from litellm.proxy.utils import PrismaClient
router = APIRouter()
_DECODED_JSON: Final = TypeAdapter(object)
class ManagedVectorStoreRow(Protocol):
@ -348,11 +350,11 @@ async def update_vector_store(
raise HTTPException(status_code=500, detail="Database not connected")
try:
update_data = data.model_dump(exclude_unset=True)
vector_store_id: Final[str] = update_data.pop("vector_store_id")
update_data = data.model_dump(exclude_unset=True, exclude={"vector_store_id"})
vector_store_id: Final = data.vector_store_id
if update_data.get("vector_store_metadata") is not None:
update_data["vector_store_metadata"] = safe_dumps(
update_data["vector_store_metadata"]
_DECODED_JSON.validate_python(update_data["vector_store_metadata"])
)
updated = await managed_vector_store_table(prisma_client).update(

View file

@ -1,6 +1,6 @@
[project]
name = "litellm-enterprise"
version = "0.1.77"
version = "0.1.79"
description = "Package for LiteLLM Enterprise features"
readme = "README.md"
requires-python = ">=3.9"
@ -26,7 +26,7 @@ required-version = ">=0.10.9"
module-root = ""
[tool.commitizen]
version = "0.1.77"
version = "0.1.79"
version_files = [
"pyproject.toml:^version",
"../pyproject.toml:litellm-enterprise==",

View file

@ -1,5 +1,5 @@
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:1d95114038f76513a9ace6fca107d5582b08c65981f81f61cb56bf7fd2ef216d
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:360896efb0753a5701c48f860e17130c205eae5a264797aa3e7ea126ee2923a7
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Checksum from https://www.pgbouncer.org/downloads/ (the Wolfi repo only carries 1.24.x)
ARG PGBOUNCER_VERSION=1.26.0
@ -20,6 +20,33 @@ RUN curl -fsSL -o pgbouncer.tar.gz "https://www.pgbouncer.org/downloads/files/${
make -j"$(nproc)" pgbouncer && \
install -m 0755 pgbouncer /usr/local/bin/pgbouncer
# ---------- Rust extension ----------
# Copies only the native build's inputs, so the fat-LTO compile stays cached across Python-only changes
FROM $LITELLM_BUILD_IMAGE AS rust-native
WORKDIR /app
USER root
COPY --from=uvbin /uv /uvx /usr/local/bin/
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
ENV UV_PYTHON_DOWNLOADS=0
COPY pyproject.toml README.md rust-toolchain.toml ./
COPY .cargo ./.cargo
COPY litellm-rust ./litellm-rust
RUN --mount=type=cache,target=/root/.cache/uv \
mkdir -p litellm/rust_bridge /wheels && \
touch litellm/__init__.py litellm/rust_bridge/__init__.py && \
uvx --python python3.13 --from maturin==1.15.0 --with puccinialin==0.1.14 \
python -c "import maturin; maturin.build_wheel('/wheels')"
# ---------- Builder ----------
FROM $LITELLM_BUILD_IMAGE AS builder
@ -32,7 +59,7 @@ COPY --from=uvbin /uv /uvx /usr/local/bin/
# instead of nodeenv downloading one whose dynamic deps may not be in Wolfi
# (e.g. Node 26.2.0 needs libatomic). Retry for transient apk.cgr.dev flakes.
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev libsndfile nodejs npm && break; \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl libssl3 openssl-dev libsndfile nodejs npm && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -66,11 +93,11 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra bedrock-realtime \
--python python3.13
# Stage 2 — copy source and install the project + workspace members.
# Stage 2 — copy source and install the workspace members.
COPY . .
RUN --mount=type=cache,target=/root/.cache/uv \
uv sync --frozen --no-default-groups --no-editable \
uv sync --frozen --no-install-project --no-default-groups --no-editable \
--extra proxy \
--extra proxy-runtime \
--group admin-mcp \
@ -79,6 +106,9 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra bedrock-realtime \
--python python3.13
RUN --mount=type=bind,from=rust-native,source=/wheels,target=/wheels \
uv pip install --python /app/.venv/bin/python --no-deps /wheels/litellm-*.whl
# PYTHONPATH=/app makes the source tree shadow the installed package, so the
# compiled Rust extension must live next to the source or it is never imported.
RUN cp "$(python -c 'import sysconfig; print(sysconfig.get_paths()["purelib"])')"/litellm/rust_bridge/_native*.so litellm/rust_bridge/
@ -95,7 +125,7 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
USER root
RUN for i in 1 2 3; do \
apk add --no-cache bash openssl tzdata python-3.13 libsndfile libatomic libevent && break; \
apk add --no-cache bash openssl libssl3 tzdata python-3.13 libsndfile libatomic libevent && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -110,7 +140,12 @@ ENV HOME=/home/nonroot \
PYTHONUNBUFFERED=1 \
PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries
COPY --from=builder --chown=nonroot:nonroot /app /app
COPY --from=builder --chown=nonroot:nonroot /app/.venv /app/.venv
COPY --from=builder --chown=nonroot:nonroot /app/litellm /app/litellm
COPY --from=builder --chown=nonroot:nonroot /app/enterprise /app/enterprise
COPY --from=builder --chown=nonroot:nonroot /app/gateway /app/gateway
COPY --from=builder --chown=nonroot:nonroot /app/docker/component_entrypoint.sh /app/docker/component_entrypoint.sh
COPY --from=builder --chown=nonroot:nonroot /app/schema.prisma /app/schema.prisma
COPY --from=builder /opt/prisma /opt/prisma
COPY --from=pgbouncer-builder /usr/local/bin/pgbouncer /usr/local/bin/pgbouncer

View file

@ -1,7 +1,7 @@
"""Path allowlist for the gateway component.
The gateway exposes the LLM data-plane surface: chat/completions, embeddings,
audio, batches, files, fine-tuning, rerank, decisions, ocr, rag, video, search, image,
audio, batches, files, fine-tuning, rerank, decisions, extraction, ocr, rag, video, search, image,
responses, vector stores, passthrough providers, realtime websockets, MCP
tool-call endpoints, and operational endpoints (/health, /metrics, and the
/debug/memory/summary read of the serving worker's RSS).
@ -64,6 +64,8 @@ GATEWAY_PATH_PREFIXES: tuple[str, ...] = (
"/decisions",
"/v1/systemone",
"/systemone",
"/v1/extraction",
"/extraction",
"/v1/ocr",
"/ocr",
"/v1/rag/",

View file

@ -70,11 +70,18 @@ proxyConfigMap:
proxy_config:
general_settings:
master_key: os.environ/PROXY_MASTER_KEY
model_list: []
```
The chart ships no models, so add them under `proxy_config.model_list` or from the Admin UI with `store_model_in_db: true`
```
proxy_config:
model_list:
- model_name: gpt-3.5-turbo
- model_name: gpt-5.6
litellm_params:
model: gpt-3.5-turbo
api_key: eXaMpLeOnLy
model: openai/gpt-5.6
api_key: os.environ/OPENAI_API_KEY
```
#### Example using existing `proxyConfigMap` instead of creating it:
@ -126,6 +133,11 @@ Set `billingMetrics.caSecretName` only when the collector is a private or test o
| `db.secret.name` | If `db.useExisting` is `true`, the name of the Kubernetes Secret that contains credentials. | `postgres` |
| `db.secret.usernameKey` | If `db.useExisting` is `true`, the name of the key within the Kubernetes Secret that holds the username for authenticating with the Postgres instance. | `username` |
| `db.secret.passwordKey` | If `db.useExisting` is `true`, the name of the key within the Kubernetes Secret that holds the password associates with the above user. | `password` |
| `db.username` | If `db.useExisting` is `true`, the database user as a plain value. When set, `db.secret.usernameKey` is not read. | `""` |
| `db.port` | If `db.useExisting` is `true` with token auth, the Postgres port used to build the connection URL. Password auth takes the port from `db.url`. | `5432` |
| `db.schema` | With token auth, the Postgres schema holding the LiteLLM tables, rendered as `DATABASE_SCHEMA` when the tables are not in `public`. Ignored with password auth, where the schema stays in `db.url` as `?schema=`. | `""` |
| `db.useIAMAuth` | If `db.useExisting` is `true` and `db.deployStandalone` is `false`, authenticate with an AWS RDS IAM token instead of a password. See [Token auth](#token-auth-to-an-existing-postgres-rds-iam-or-microsoft-entra-id). | `false` |
| `db.useAzureEntraAuth` | If `db.useExisting` is `true` and `db.deployStandalone` is `false`, authenticate with a Microsoft Entra ID token instead of a password. Mutually exclusive with `db.useIAMAuth`. | `false` |
| `db.useStackgresOperator` | Not yet implemented. | `false` |
| `db.deployStandalone` | Deploy a standalone, single instance deployment of Postgres, using the Bitnami postgresql chart. This is useful for getting started but doesn't provide HA or (by default) data backups. | `true` |
| `postgresql.*` | If `db.deployStandalone` is `true`, configuration passed to the Bitnami postgresql chart. See the [Bitnami Documentation](https://github.com/bitnami/charts/tree/main/bitnami/postgresql) for full configuration details. See [values.yaml](./values.yaml) for the default configuration. | See [values.yaml](./values.yaml) |
@ -156,6 +168,22 @@ data:
type: Opaque
```
#### Token auth to an existing Postgres (RDS IAM or Microsoft Entra ID)
When the database accepts short-lived tokens instead of a password (AWS RDS or Aurora with IAM database authentication, or Azure Database for PostgreSQL with Microsoft Entra ID), there is no password to put in a Secret. Turn on `db.useIAMAuth` or `db.useAzureEntraAuth` and the chart stops referencing `db.secret.passwordKey` and `db.url`. It sets `IAM_TOKEN_DB_AUTH` (or `AZURE_POSTGRESQL_AUTH`) and emits `DATABASE_HOST`, `DATABASE_PORT`, `DATABASE_USERNAME` and `DATABASE_NAME` for the proxy, the migrations job and the collector sidecar. Each of them mints its own token at startup and builds `DATABASE_URL` from those parts, and the in-container pgbouncer (`db.connectionPool`) refreshes the token before it expires.
```yaml
db:
useExisting: true
deployStandalone: false
useIAMAuth: true
endpoint: litellm.cluster-abc123.us-west-1.rds.amazonaws.com
database: litellm
username: litellm_iam # the role granted rds_iam; no db.secret needed
```
The pod needs cloud credentials for the token: on EKS use IRSA or Pod Identity on the chart's `serviceAccount` (and `migrationJob.serviceAccountName` when the Helm hook runs before the ServiceAccount exists), on AKS use workload identity. Set `AWS_RDS_REGION` under `envVars` when the RDS hostname is not a standard regional endpoint (the token must be signed for the region the database lives in). TLS options that would otherwise go in `db.url` are set as `DATABASE_SSLMODE` and `DATABASE_SSLROOTCERT` under `envVars`. Leaving `db.username` empty reads the user from `db.secret.usernameKey` as before.
#### Examples for `environmentSecrets` and `environemntConfigMaps`
```yaml

View file

@ -162,6 +162,77 @@ taken before the change, which by that point no longer exists.
{{- end -}}
{{- end -}}
{{/*
Whether the existing Postgres is reached with a short-lived token (AWS RDS IAM
or Microsoft Entra ID) instead of a password.
*/}}
{{- define "litellm.db.tokenAuth" -}}
{{- if or .Values.db.useIAMAuth .Values.db.useAzureEntraAuth -}}true{{- end -}}
{{- end -}}
{{- define "litellm.validateDbTokenAuth" -}}
{{- if and .Values.db.useIAMAuth .Values.db.useAzureEntraAuth -}}
{{- fail "db.useIAMAuth and db.useAzureEntraAuth are mutually exclusive: the database password can only come from one token source" -}}
{{- end -}}
{{- if and (include "litellm.db.tokenAuth" .) (or (not .Values.db.useExisting) .Values.db.deployStandalone) -}}
{{- fail "db.useIAMAuth and db.useAzureEntraAuth need db.useExisting: true and db.deployStandalone: false (token auth targets an existing RDS or Azure Database for PostgreSQL; the bundled postgresql subchart only supports password auth)" -}}
{{- end -}}
{{- end -}}
{{/*
Database env for an existing Postgres, shared by the proxy container, the
collector sidecar and the migrations Job. Under token auth every entrypoint
mints its own token and assembles DATABASE_URL itself, so db.url is left out.
*/}}
{{- define "litellm.existingDbEnv" -}}
{{- $tokenAuth := include "litellm.db.tokenAuth" . -}}
- name: DATABASE_USERNAME
{{- if .Values.db.username }}
value: {{ .Values.db.username | quote }}
{{- else }}
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.usernameKey }}
{{- end }}
{{- if .Values.db.useIAMAuth }}
- name: IAM_TOKEN_DB_AUTH
value: "true"
{{- else if .Values.db.useAzureEntraAuth }}
- name: AZURE_POSTGRESQL_AUTH
value: "true"
{{- else }}
- name: DATABASE_PASSWORD
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.passwordKey }}
{{- end }}
- name: DATABASE_HOST
{{- if .Values.db.secret.endpointKey }}
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.endpointKey }}
{{- else }}
value: {{ .Values.db.endpoint }}
{{- end }}
{{- if $tokenAuth }}
- name: DATABASE_PORT
value: {{ .Values.db.port | quote }}
{{- end }}
- name: DATABASE_NAME
value: {{ .Values.db.database }}
{{- if and $tokenAuth .Values.db.schema }}
- name: DATABASE_SCHEMA
value: {{ .Values.db.schema | quote }}
{{- end }}
{{- if not $tokenAuth }}
- name: DATABASE_URL
value: {{ .Values.db.url | quote }}
{{- end }}
{{- end -}}
{{/*
Environment shared by the proxy container and the opt-in collector sidecar:
database, pgbouncer, master key, redis, user envVars. Both containers must see
@ -169,6 +240,7 @@ the same DATABASE_URL and REDIS_* so the sidecar reaches the pod's pgbouncer
and the same spend transaction buffer.
*/}}
{{- define "litellm.proxyEnv" -}}
{{- include "litellm.validateDbTokenAuth" . -}}
- name: HOST
value: "{{ .Values.listen | default "0.0.0.0" }}"
- name: PORT
@ -189,29 +261,7 @@ and the same spend transaction buffer.
- name: DATABASE_NAME
value: litellm
{{- else if .Values.db.useExisting }}
- name: DATABASE_USERNAME
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.usernameKey }}
- name: DATABASE_PASSWORD
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.passwordKey }}
- name: DATABASE_HOST
{{- if .Values.db.secret.endpointKey }}
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.endpointKey }}
{{- else }}
value: {{ .Values.db.endpoint }}
{{- end }}
- name: DATABASE_NAME
value: {{ .Values.db.database }}
- name: DATABASE_URL
value: {{ .Values.db.url | quote }}
{{ include "litellm.existingDbEnv" . }}
{{- end }}
{{- if and .Values.db.useExisting .Values.db.readReplicaUrl .Values.db.secret.readReplicaEndpointKey (not .Values.db.secret.readReplicaUrlKey) }}
- name: DATABASE_READER_HOST

View file

@ -68,30 +68,9 @@ spec:
{{- end }}
workingDir: "/app"
env:
{{- include "litellm.validateDbTokenAuth" . -}}
{{- if .Values.db.useExisting }}
- name: DATABASE_USERNAME
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.usernameKey }}
- name: DATABASE_PASSWORD
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.passwordKey }}
- name: DATABASE_HOST
{{- if .Values.db.secret.endpointKey }}
valueFrom:
secretKeyRef:
name: {{ .Values.db.secret.name }}
key: {{ .Values.db.secret.endpointKey }}
{{- else }}
value: {{ .Values.db.endpoint }}
{{- end }}
- name: DATABASE_NAME
value: {{ .Values.db.database }}
- name: DATABASE_URL
value: {{ .Values.db.url | quote }}
{{- include "litellm.existingDbEnv" . | nindent 12 }}
{{- else if .Values.db.deployStandalone }}
- name: DATABASE_URL
value: postgresql://{{ .Values.postgresql.auth.username }}:{{ .Values.postgresql.auth.password }}@{{ .Release.Name }}-postgresql/{{ .Values.postgresql.auth.database }}

View file

@ -0,0 +1,288 @@
suite: test existing database token auth
templates:
- deployment.yaml
- migrations-job.yaml
- configmap-litellm.yaml
tests:
- it: should keep password auth from the secret when no token auth is set
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_USERNAME
valueFrom:
secretKeyRef:
name: postgres
key: username
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PASSWORD
valueFrom:
secretKeyRef:
name: postgres
key: password
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_URL
value: postgresql://$(DATABASE_USERNAME):$(DATABASE_PASSWORD)@$(DATABASE_HOST)/$(DATABASE_NAME)
- notContains:
path: spec.template.spec.containers[0].env
content:
name: IAM_TOKEN_DB_AUTH
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: AZURE_POSTGRESQL_AUTH
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PORT
any: true
- it: should render RDS IAM auth without a password secret on the proxy and the collector sidecar
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
db.useIAMAuth: true
db.username: litellm_iam
db.endpoint: writer.cluster-abc.us-west-1.rds.amazonaws.com
db.port: 5433
collector.enabled: true
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: IAM_TOKEN_DB_AUTH
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_USERNAME
value: litellm_iam
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_HOST
value: writer.cluster-abc.us-west-1.rds.amazonaws.com
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PORT
value: "5433"
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_NAME
value: litellm
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PASSWORD
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_URL
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: AZURE_POSTGRESQL_AUTH
any: true
- contains:
path: spec.template.spec.containers[1].env
content:
name: IAM_TOKEN_DB_AUTH
value: "true"
- notContains:
path: spec.template.spec.containers[1].env
content:
name: DATABASE_PASSWORD
any: true
- it: should give the migrations job the same token auth env as the proxy
template: migrations-job.yaml
set:
migrationJob.enabled: true
db.useExisting: true
db.deployStandalone: false
db.useIAMAuth: true
db.username: litellm_iam
db.endpoint: writer.cluster-abc.us-west-1.rds.amazonaws.com
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: IAM_TOKEN_DB_AUTH
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_USERNAME
value: litellm_iam
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_HOST
value: writer.cluster-abc.us-west-1.rds.amazonaws.com
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PORT
value: "5432"
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PASSWORD
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_URL
any: true
- it: should carry db.schema as DATABASE_SCHEMA under token auth and omit it when empty
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
db.useIAMAuth: true
db.username: litellm_iam
db.schema: tenant1
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_SCHEMA
value: tenant1
- it: should leave the schema to db.url under password auth even when db.schema is set
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
db.schema: tenant1
db.url: postgresql://$(DATABASE_USERNAME):$(DATABASE_PASSWORD)@$(DATABASE_HOST)/$(DATABASE_NAME)?schema=tenant1
asserts:
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_SCHEMA
any: true
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_URL
value: postgresql://$(DATABASE_USERNAME):$(DATABASE_PASSWORD)@$(DATABASE_HOST)/$(DATABASE_NAME)?schema=tenant1
- it: should not emit DATABASE_SCHEMA when db.schema is empty
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
db.useIAMAuth: true
asserts:
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_SCHEMA
any: true
- it: should read the user from the secret when db.username is empty under token auth
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
db.useIAMAuth: true
db.secret.name: rds-identity
db.secret.usernameKey: user
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_USERNAME
valueFrom:
secretKeyRef:
name: rds-identity
key: user
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PASSWORD
any: true
- it: should render Entra auth with AZURE_POSTGRESQL_AUTH and no password
template: deployment.yaml
set:
db.useExisting: true
db.deployStandalone: false
db.useAzureEntraAuth: true
db.username: litellm@example.onmicrosoft.com
db.endpoint: litellm.postgres.database.azure.com
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: AZURE_POSTGRESQL_AUTH
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_USERNAME
value: litellm@example.onmicrosoft.com
- notContains:
path: spec.template.spec.containers[0].env
content:
name: IAM_TOKEN_DB_AUTH
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_PASSWORD
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: DATABASE_URL
any: true
- it: should fail when both token sources are enabled
template: deployment.yaml
set:
migrationJob.enabled: false
db.useExisting: true
db.deployStandalone: false
db.useIAMAuth: true
db.useAzureEntraAuth: true
asserts:
- failedTemplate:
errorMessage: "db.useIAMAuth and db.useAzureEntraAuth are mutually exclusive: the database password can only come from one token source"
- it: should fail when token auth is set while the bundled postgres is still deployed
template: deployment.yaml
set:
migrationJob.enabled: false
db.useExisting: true
db.useIAMAuth: true
asserts:
- failedTemplate:
errorPattern: "need db.useExisting: true and db.deployStandalone: false"
- it: should fail when token auth is set without useExisting
template: migrations-job.yaml
set:
migrationJob.enabled: true
db.deployStandalone: false
db.useAzureEntraAuth: true
asserts:
- failedTemplate:
errorPattern: "need db.useExisting: true and db.deployStandalone: false"

View file

@ -270,6 +270,32 @@ tests:
- matchRegex:
path: data["config.yaml"]
pattern: '(?m)^\s*health_check_details:\s*false$'
- it: should not render any placeholder models when the user sets proxy_config without model_list
template: configmap-litellm.yaml
set:
proxy_config.general_settings.store_model_in_db: true
asserts:
- notMatchRegex:
path: data["config.yaml"]
pattern: 'model_name:'
- matchRegex:
path: data["config.yaml"]
pattern: 'store_model_in_db: true'
- matchRegex:
path: data["config.yaml"]
pattern: 'master_key: os.environ/PROXY_MASTER_KEY'
- it: should render a user-supplied model_list as the only models
template: configmap-litellm.yaml
set:
proxy_config.model_list:
- model_name: my-model
litellm_params:
model: openai/my-model
api_key: os.environ/OPENAI_API_KEY
asserts:
- matchRegex:
path: data["config.yaml"]
pattern: 'model_list:\n- litellm_params:\n api_key: os.environ/OPENAI_API_KEY\n model: openai/my-model\n model_name: my-model\n$'
- it: should allow overriding liveness, readiness, and startup probes
template: deployment.yaml
set:

View file

@ -180,17 +180,7 @@ proxyConfigMap:
# Examples: https://github.com/BerriAI/litellm/tree/main/litellm/proxy/example_config_yaml
# Reference: https://docs.litellm.ai/docs/proxy/configs
proxy_config:
model_list:
# At least one model must exist for the proxy to start.
- model_name: gpt-3.5-turbo
litellm_params:
model: gpt-3.5-turbo
api_key: eXaMpLeOnLy
- model_name: fake-openai-endpoint
litellm_params:
model: openai/fake
api_key: fake-key
api_base: https://exampleopenaiendpoint-production.up.railway.app/
model_list: []
general_settings:
master_key: os.environ/PROXY_MASTER_KEY
@ -376,6 +366,30 @@ db:
endpoint: localhost
database: litellm
url: postgresql://$(DATABASE_USERNAME):$(DATABASE_PASSWORD)@$(DATABASE_HOST)/$(DATABASE_NAME)
# Authenticate with a short-lived token instead of a password. useIAMAuth is
# AWS RDS / Aurora IAM database authentication (the pod needs AWS credentials,
# e.g. IRSA or EKS Pod Identity, and the database user needs GRANT rds_iam);
# useAzureEntraAuth is Azure Database for PostgreSQL with a Microsoft Entra ID
# identity (workload identity). Mutually exclusive, and both need
# useExisting: true with deployStandalone: false. The chart then sets
# IAM_TOKEN_DB_AUTH or AZURE_POSTGRESQL_AUTH, emits DATABASE_HOST, DATABASE_PORT,
# DATABASE_USERNAME and DATABASE_NAME and skips DATABASE_PASSWORD and db.url:
# the proxy, the migrations job and the collector sidecar each mint their own
# token and build DATABASE_URL from those parts. Set DATABASE_SSLMODE /
# DATABASE_SSLROOTCERT under envVars for TLS options you would otherwise put
# in db.url.
useIAMAuth: false
useAzureEntraAuth: false
# Postgres port, used to build the token-auth URL (password auth takes the
# port from db.url)
port: 5432
# Postgres schema for the LiteLLM tables, rendered as DATABASE_SCHEMA under
# token auth only (password auth carries it as ?schema= in db.url)
schema: ""
# Database user as a plain value. Empty means read it from
# db.secret.usernameKey. With token auth the user name is the only credential
# the chart needs, so setting it here removes the need for db.secret entirely.
username: ""
secret:
name: postgres
usernameKey: username

View file

@ -475,6 +475,27 @@ harmless no-op for the Job and authoritative for the app pods.
{{- end }}
{{- end }}
{{- end }}
{{- $sslCertReqs := $root.Values.redis.sslCertReqs }}
{{- if and $sslCertReqs (not (has $sslCertReqs (list "none" "optional" "required"))) }}
{{- fail (printf "redis.sslCertReqs must be one of none, optional, required (got %q)" $sslCertReqs) }}
{{- end }}
{{- if and $urlKey (or $root.Values.redis.ssl $sslCertReqs $root.Values.redis.sslCaCerts) }}
{{- fail "redis.ssl, redis.sslCertReqs and redis.sslCaCerts cannot be combined with redis.passwordSecret.urlKey, use a rediss:// URL with ssl_cert_reqs / ssl_ca_certs query parameters instead" }}
{{- end }}
{{- if or $root.Values.redis.host $hostKey }}
{{- if $root.Values.redis.ssl }}
- name: REDIS_SSL
value: "true"
{{- end }}
{{- if $sslCertReqs }}
- name: REDIS_SSL_CERT_REQS
value: {{ $sslCertReqs | quote }}
{{- end }}
{{- if $root.Values.redis.sslCaCerts }}
- name: REDIS_SSL_CA_CERTS
value: {{ $root.Values.redis.sslCaCerts | quote }}
{{- end }}
{{- end }}
{{- with $component.extraEnv }}
{{ toYaml . }}
{{- end }}

View file

@ -3,6 +3,7 @@ templates:
- gateway/deployment.yaml
- gateway/configmap.yaml
- backend/deployment.yaml
- migrations-job.yaml
values:
- ./values/required.yaml
tests:
@ -320,3 +321,168 @@ tests:
asserts:
- failedTemplate:
errorMessage: "redis.passwordSecret.name is required when hostKey or urlKey is set"
- it: redis omits TLS env vars by default
template: gateway/deployment.yaml
set:
redis.host: redis.example.com
asserts:
- notContains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CERT_REQS
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CA_CERTS
any: true
- it: redis ssl false omits REDIS_SSL
template: gateway/deployment.yaml
set:
redis.host: redis.example.com
redis.ssl: false
redis.sslCertReqs: required
asserts:
- notContains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
any: true
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CERT_REQS
value: "required"
- it: gateway renders redis TLS env vars
template: gateway/deployment.yaml
set:
redis.host: redis.example.com
redis.ssl: true
redis.sslCertReqs: none
redis.sslCaCerts: /etc/redis-tls/ca.crt
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CERT_REQS
value: "none"
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CA_CERTS
value: "/etc/redis-tls/ca.crt"
- it: backend renders redis TLS env vars
template: backend/deployment.yaml
set:
redis.host: redis.example.com
redis.ssl: true
redis.sslCertReqs: required
redis.sslCaCerts: /etc/redis-tls/ca.crt
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CERT_REQS
value: "required"
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CA_CERTS
value: "/etc/redis-tls/ca.crt"
- it: migration job renders redis TLS env vars
template: migrations-job.yaml
set:
db.deployStandalone: false
db.useExisting: true
db.endpoint: postgres.example.com
db.secret.name: pg-secret
redis.host: redis.example.com
redis.ssl: true
redis.sslCertReqs: optional
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CERT_REQS
value: "optional"
- it: redis TLS values reject a secret-backed URL
template: gateway/deployment.yaml
set:
redis.passwordSecret.name: redis-secret
redis.passwordSecret.urlKey: connection-url
redis.sslCaCerts: /etc/redis-tls/ca.crt
asserts:
- failedTemplate:
errorMessage: "redis.ssl, redis.sslCertReqs and redis.sslCaCerts cannot be combined with redis.passwordSecret.urlKey, use a rediss:// URL with ssl_cert_reqs / ssl_ca_certs query parameters instead"
- it: hostKey mode renders redis TLS env vars
template: gateway/deployment.yaml
set:
redis.passwordSecret.name: redis-secret
redis.passwordSecret.hostKey: hostname
redis.ssl: true
redis.sslCaCerts: /etc/redis-tls/ca.crt
asserts:
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
value: "true"
- contains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CA_CERTS
value: "/etc/redis-tls/ca.crt"
- it: redis TLS env vars are omitted when redis is not configured
template: gateway/deployment.yaml
set:
redis.ssl: true
redis.sslCertReqs: none
asserts:
- notContains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL
any: true
- notContains:
path: spec.template.spec.containers[0].env
content:
name: REDIS_SSL_CERT_REQS
any: true
- it: redis sslCertReqs rejects unknown values
template: gateway/deployment.yaml
set:
redis.host: redis.example.com
redis.ssl: true
redis.sslCertReqs: strict
asserts:
- failedTemplate:
errorMessage: 'redis.sslCertReqs must be one of none, optional, required (got "strict")'

View file

@ -275,10 +275,21 @@ database:
# self-hosted Redis Cluster). The chart emits REDIS_CLUSTER_NODES from
# `host` / `port` as the single seed; the cluster client discovers the
# remaining nodes from CLUSTER SLOTS at startup.
#
# For a TLS Redis endpoint set `ssl: true`, which the chart emits as REDIS_SSL.
# `sslCertReqs` (none / optional / required) becomes REDIS_SSL_CERT_REQS and
# `sslCaCerts` (path to a CA bundle mounted in the pod, e.g. via
# `extraVolumeMounts`) becomes REDIS_SSL_CA_CERTS. All three are omitted when
# unset. They apply to `host` / `hostKey` mode only. With `passwordSecret.urlKey`
# the proxy reads TLS settings from the URL itself, so use `rediss://` and the
# `ssl_cert_reqs` / `ssl_ca_certs` query parameters there instead.
redis:
cluster: false
host: ""
port: 6379
ssl: false
sslCertReqs: ""
sslCaCerts: ""
passwordSecret:
name: "" # Leave empty for auth-less Redis
passwordKey: password

View file

@ -0,0 +1,50 @@
import functools
import os
import re
from collections.abc import Callable
from typing import Final
from urllib.parse import unquote, urlsplit
_REDACTED: Final = "REDACTED"
_PASSWORD_QUERY_KEYS: Final = frozenset(("password", "sslpassword"))
@functools.cache
def _secret_shape_redactor() -> Callable[[str], str]:
try:
from litellm._logging import redact_secrets
except ImportError:
return lambda text: text
return redact_secrets
def _url_passwords(url: str) -> frozenset[str]:
try:
parts: Final = urlsplit(url)
except ValueError:
return frozenset()
query_pairs: Final = tuple(pair.partition("=") for pair in parts.query.split("&"))
raw_query_passwords: Final = tuple(
value for key, separator, value in query_pairs if separator and key.lower() in _PASSWORD_QUERY_KEYS
)
raw_passwords: Final = ((parts.password,) if parts.password else ()) + raw_query_passwords
return frozenset(password for password in raw_passwords + tuple(map(unquote, raw_passwords)) if password)
def _configured_database_passwords() -> frozenset[str]:
database_url: Final = os.getenv("DATABASE_URL")
direct_url: Final = os.getenv("DIRECT_URL")
database_passwords: Final = _url_passwords(database_url) if database_url else frozenset()
direct_passwords: Final = _url_passwords(direct_url) if direct_url else frozenset()
return database_passwords | direct_passwords
def _redact_credentials(text: str) -> str:
"""Mask configured database passwords before passing the text to LiteLLM redaction."""
passwords: Final = sorted(_configured_database_passwords(), key=len, reverse=True)
alternation: Final = "|".join(re.escape(password) for password in passwords)
password_pattern: Final = (
re.compile(rf"(?<![A-Za-z0-9_])(?:{alternation})(?![A-Za-z0-9_])", re.IGNORECASE) if passwords else None
)
result: Final = password_pattern.sub(_REDACTED, text) if password_pattern is not None else text
return _secret_shape_redactor()(result)

View file

@ -0,0 +1,631 @@
"""Read-only checks that the rows already in the database can take the pending schema
changes, run before `prisma migrate deploy` and before each request-log index build.
Postgres has a short list of schema statements that fail because of existing rows: a
btree index whose entry for some row would not fit in a third of a page (2704 bytes on
the default 8 KB page, counting every indexed column), UNIQUE (duplicate values), SET NOT NULL and ADD COLUMN NOT NULL with no default (nulls, any row at all),
FOREIGN KEY (orphans) and CHECK (rows failing the expression). Each pending statement
of one of those kinds becomes one count query derived from the statement text and the
column types in the catalog. A count above zero names the table, the column, the
statement it blocks and the SQL that clears it, and the migration is refused before
anything is applied. Every other statement a migration can contain cannot fail on
existing data and gets no query. A statement the derivation cannot read gets none
either, so the migration keeps today's behavior for it.
"""
import os
import re
from collections.abc import Callable, Iterator
from dataclasses import dataclass
from functools import reduce
from itertools import accumulate
from pathlib import Path
from types import MappingProxyType
from typing import TYPE_CHECKING, Final, Literal
from litellm_proxy_extras._logging import logger
if TYPE_CHECKING:
import psycopg
from psycopg import sql
DEFAULT_BLOCK_SIZE: Final = 8192
INDEX_TUPLE_HEADER_BYTES: Final = 8
NULL_BITMAP_BYTES: Final = 8
VARLENA_HEADER_BYTES: Final = 4
SHORT_VARLENA_MAX_BYTES: Final = 126
_PAGE_HEADER_WITH_THREE_LINE_POINTERS_BYTES: Final = 40
_BTREE_PAGE_OPAQUE_BYTES: Final = 16
_HEAP_TID_BYTES: Final = 8
_ALIGNMENT: Final = MappingProxyType({"c": 1, "s": 2, "i": 4, "d": 8})
_OCTET_LENGTH_TYPES: Final = frozenset({"text", "varchar", "bpchar", "bytea", "name"})
MIGRATION_DATA_CHECK_TIMEOUT_ENV_VAR: Final = "LITELLM_MIGRATION_DATA_CHECK_TIMEOUT"
DEFAULT_MIGRATION_DATA_CHECK_TIMEOUT_SECONDS: Final = 600.0
REFUSAL: Final = "Existing rows would make a pending schema change fail:"
Kind = Literal["too_long", "duplicates", "nulls", "any_row", "orphans", "check_violations"]
@dataclass(frozen=True, slots=True)
class Check:
kind: Kind
table: str
columns: tuple[str, ...]
statement: str
source: str
index: str | None = None
referenced_table: str | None = None
referenced_columns: tuple[str, ...] = ()
expression: str | None = None
@dataclass(frozen=True, slots=True)
class BlockedChange:
check: Check
rows: int
detail: str
@dataclass(frozen=True, slots=True)
class CouldNotVerify:
check: Check
timeout_seconds: float
Finding = BlockedChange | CouldNotVerify
@dataclass(frozen=True, slots=True)
class ColumnType:
"""One column as `pg_attribute` and `pg_type` describe it: `length` is `attlen` (-1 for a
variable-length value), `align` is `attalign` (c, s, i or d), `typmod` is `atttypmod`."""
type_name: str
length: int
align: str
typmod: int
@property
def is_variable_length(self) -> bool:
return self.length == -1
@property
def can_overflow_btree(self) -> bool:
"""Whether one value of this column alone can push a btree entry past the limit: a
variable-length type with no declared cap, or a cap of more bytes than the limit."""
if not self.is_variable_length:
return False
return self.typmod == -1 or (self.typmod - VARLENA_HEADER_BYTES) * 4 > btree_max_entry_bytes(DEFAULT_BLOCK_SIZE)
ColumnTypeLookup = Callable[[str, str], ColumnType | None]
_IDENT: Final = r'"?(\w+)"?'
_QUALIFIED: Final = r'(?:"?\w+"?\s*\.\s*)?"?(\w+)"?'
_CREATE_INDEX: Final = re.compile(
rf"^\s*CREATE\s+(UNIQUE\s+)?INDEX\s+(?:CONCURRENTLY\s+)?(?:IF\s+NOT\s+EXISTS\s+)?{_IDENT}\s+ON\s+(?:ONLY\s+)?"
rf"{_QUALIFIED}\s*(?:USING\s+(\w+)\s*)?\((.*?)\)",
re.IGNORECASE | re.DOTALL,
)
_ALTER_TABLE: Final = re.compile(
rf"^\s*ALTER\s+TABLE\s+(?:IF\s+EXISTS\s+)?(?:ONLY\s+)?{_QUALIFIED}\s+(.*)$", re.IGNORECASE | re.DOTALL
)
_SET_NOT_NULL: Final = re.compile(rf"^ALTER\s+(?:COLUMN\s+)?{_IDENT}\s+SET\s+NOT\s+NULL$", re.IGNORECASE)
_ADD_COLUMN: Final = re.compile(
rf"^ADD\s+(?:COLUMN\s+)?(?:IF\s+NOT\s+EXISTS\s+)?(?!(?:CONSTRAINT|FOREIGN|UNIQUE|CHECK|PRIMARY|EXCLUDE)\b){_IDENT}\s+(.*)$",
re.IGNORECASE | re.DOTALL,
)
_CONSTRAINT: Final = rf"^ADD\s+(?:CONSTRAINT\s+{_IDENT}\s+)?"
_ADD_FOREIGN_KEY: Final = re.compile(
rf"{_CONSTRAINT}FOREIGN\s+KEY\s*\(([^)]*)\)\s*REFERENCES\s+{_QUALIFIED}\s*\(([^)]*)\)(.*)$",
re.IGNORECASE | re.DOTALL,
)
_ADD_UNIQUE: Final = re.compile(rf"{_CONSTRAINT}UNIQUE\s*\(([^)]*)\)", re.IGNORECASE | re.DOTALL)
_ADD_CHECK: Final = re.compile(rf"{_CONSTRAINT}CHECK\s*\((.*)\)\s*(NOT\s+VALID)?\s*$", re.IGNORECASE | re.DOTALL)
_PLAIN_COLUMN: Final = re.compile(rf"^{_IDENT}(?:\s+(?:ASC|DESC|NULLS\s+(?:FIRST|LAST)))*$", re.IGNORECASE)
_COMMENT: Final = re.compile(r"--[^\n]*")
_NOT_NULL: Final = re.compile(r"\bNOT\s+NULL\b", re.IGNORECASE)
_HAS_DEFAULT: Final = re.compile(r"\bDEFAULT\b|\bGENERATED\b", re.IGNORECASE)
def migration_data_check_timeout_seconds() -> float:
"""Seconds one count query may run before the preflight gives up on verifying it."""
raw: Final = os.getenv(MIGRATION_DATA_CHECK_TIMEOUT_ENV_VAR)
if raw is None:
return DEFAULT_MIGRATION_DATA_CHECK_TIMEOUT_SECONDS
try:
return float(raw)
except ValueError:
logger.warning(
"%s=%r is not a number, using %s",
MIGRATION_DATA_CHECK_TIMEOUT_ENV_VAR,
raw,
DEFAULT_MIGRATION_DATA_CHECK_TIMEOUT_SECONDS,
)
return DEFAULT_MIGRATION_DATA_CHECK_TIMEOUT_SECONDS
def statements(migration_sql: str) -> tuple[str, ...]:
"""The statements of one migration file, comments stripped, empty ones dropped."""
parts: Final = _COMMENT.sub("", migration_sql).split(";")
return tuple(part.strip() for part in parts if part.strip())
def _split_top_level(text: str) -> tuple[str, ...]:
"""Split on the commas outside parentheses."""
depth: Final = accumulate((1 if char == "(" else -1 if char == ")" else 0) for char in text)
cuts: Final = tuple(
position for position, (char, level) in enumerate(zip(text, depth, strict=True)) if char == "," and level == 0
)
starts: Final = (0, *(cut + 1 for cut in cuts))
stops: Final = (*cuts, len(text))
pieces: Final = tuple(text[start:stop].strip() for start, stop in zip(starts, stops, strict=True))
return tuple(piece for piece in pieces if piece)
def _plain_columns(column_list: str) -> tuple[str, ...] | None:
"""The column names of an index or constraint column list, None when one entry is an
expression the derivation cannot turn into a count."""
matches: Final = tuple(_PLAIN_COLUMN.match(entry) for entry in _split_top_level(column_list))
if any(match is None for match in matches):
return None
return tuple(match.group(1) for match in matches if match is not None)
def _index_checks(match: "re.Match[str]", statement_text: str, source: str) -> Iterator[Check]:
unique, name, table, method, column_list = match.groups()
columns: Final = _plain_columns(column_list)
if columns is None:
return
label: Final = f'CREATE {"UNIQUE " if unique else ""}INDEX "{name}"'
if method is None or method.lower() == "btree":
yield Check("too_long", table, columns, label, source, index=name)
if unique:
yield Check("duplicates", table, columns, label, source, index=name)
def _alter_table_checks(table: str, action: str, source: str) -> Iterator[Check]:
if (set_not_null := _SET_NOT_NULL.match(action)) is not None:
yield Check("nulls", table, (set_not_null.group(1),), f'ALTER TABLE "{table}" {action}', source)
return
if (add_column := _ADD_COLUMN.match(action)) is not None:
column, body = add_column.groups()
if _NOT_NULL.search(body) is not None and _HAS_DEFAULT.search(body) is None:
yield Check(
"any_row", table, (column,), f'ALTER TABLE "{table}" ADD COLUMN "{column}" ... NOT NULL', source
)
return
if (foreign_key := _ADD_FOREIGN_KEY.match(action)) is not None:
constraint, columns, referenced_table, referenced_columns, tail = foreign_key.groups()
own: Final = _plain_columns(columns)
theirs: Final = _plain_columns(referenced_columns)
if own is None or theirs is None or len(own) != len(theirs) or "NOT VALID" in tail.upper():
return
yield Check(
"orphans",
table,
own,
_constraint_label(table, constraint, "FOREIGN KEY"),
source,
referenced_table=referenced_table,
referenced_columns=theirs,
)
return
if (unique := _ADD_UNIQUE.match(action)) is not None:
constraint, columns = unique.groups()
unique_columns: Final = _plain_columns(columns)
if unique_columns is not None:
yield Check("duplicates", table, unique_columns, _constraint_label(table, constraint, "UNIQUE"), source)
return
if (check := _ADD_CHECK.match(action)) is not None:
constraint, expression, not_valid = check.groups()
if not_valid is None:
yield Check(
"check_violations",
table,
(),
_constraint_label(table, constraint, "CHECK"),
source,
expression=expression,
)
def _constraint_label(table: str, constraint: str | None, kind: str) -> str:
named: Final = f'CONSTRAINT "{constraint}" ' if constraint else ""
return f'ALTER TABLE "{table}" ADD {named}{kind}'
def checks_for_statement(statement_text: str, source: str) -> tuple[Check, ...]:
"""The count queries one SQL statement needs before it can be applied, empty for a
statement that cannot fail on existing rows or that the derivation cannot read."""
if (index := _CREATE_INDEX.match(statement_text)) is not None:
return tuple(_index_checks(index, statement_text, source))
if (alter := _ALTER_TABLE.match(statement_text)) is not None:
table, actions = alter.groups()
return tuple(
check for action in _split_top_level(actions) for check in _alter_table_checks(table, action, source)
) # comprehension-ok: flattening the per-action checks of one statement
return ()
def checks_for_migration(name: str, migration_sql: str) -> tuple[Check, ...]:
return tuple(
check
for statement_text in statements(migration_sql)
for check in checks_for_statement(statement_text, f"migration {name}")
) # comprehension-ok: flattening the per-statement checks of one migration
def checks_for_index(table: str, name: str, columns: tuple[str, ...]) -> tuple[Check, ...]:
"""The entry-size check a request-log index needs before `CREATE INDEX CONCURRENTLY`."""
return (Check("too_long", table, columns, f'CREATE INDEX "{name}"', "request-log index", index=name),)
def _applied_migrations(connection: "psycopg.Connection[tuple[object, ...]]", schema: str) -> frozenset[str]:
import psycopg
from psycopg import sql
try:
rows: Final = connection.execute(
sql.SQL("SELECT migration_name FROM {} WHERE finished_at IS NOT NULL AND rolled_back_at IS NULL").format(
sql.Identifier(schema, "_prisma_migrations")
)
).fetchall()
except psycopg.errors.UndefinedTable:
return frozenset()
return frozenset(str(row[0]) for row in rows)
def pending_migrations(
connection: "psycopg.Connection[tuple[object, ...]]", schema: str, migrations_dir: Path
) -> tuple[tuple[str, str], ...]:
"""(name, sql) of every shipped migration the ledger has not finished, oldest first.
Without a ledger every migration is pending, and the checks skip tables that do not exist yet."""
applied: Final = _applied_migrations(connection, schema)
folders: Final = tuple(folder for folder in sorted(migrations_dir.iterdir()) if folder.is_dir())
return tuple(
(folder.name, (folder / "migration.sql").read_text())
for folder in folders
if folder.name not in applied and (folder / "migration.sql").is_file()
)
def _int(value: object, default: int) -> int:
return value if isinstance(value, int) else default
def column_type_lookup(connection: "psycopg.Connection[tuple[object, ...]]", schema: str) -> ColumnTypeLookup:
def lookup(table: str, column: str) -> ColumnType | None:
row: Final = connection.execute(
"SELECT t.typname, a.attlen, a.attalign, a.atttypmod FROM pg_attribute a "
"JOIN pg_class c ON c.oid = a.attrelid JOIN pg_namespace n ON n.oid = c.relnamespace "
"JOIN pg_type t ON t.oid = a.atttypid "
"WHERE n.nspname = %s AND c.relname = %s AND a.attname = %s AND a.attnum > 0 AND NOT a.attisdropped",
(schema, table, column),
).fetchone()
if row is None:
return None
return ColumnType(str(row[0]), _int(row[1], -1), str(row[2]), _int(row[3], -1))
return lookup
def _column_types(check: Check, column_type: ColumnTypeLookup) -> tuple[ColumnType, ...] | None:
"""The catalog types of the check's columns, None when one of them is not there yet."""
types: Final = tuple(column_type(check.table, column) for column in check.columns)
if any(column is None for column in types):
return None
return tuple(column for column in types if column is not None)
def _table_exists(connection: "psycopg.Connection[tuple[object, ...]]", schema: str, table: str) -> bool:
row: Final = connection.execute(
"SELECT 1 FROM information_schema.tables WHERE table_schema = %s AND table_name = %s", (schema, table)
).fetchone()
return row is not None
def _index_is_valid(connection: "psycopg.Connection[tuple[object, ...]]", schema: str, index: str) -> bool:
row: Final = connection.execute(
"SELECT x.indisvalid FROM pg_index x JOIN pg_class i ON i.oid = x.indexrelid "
"JOIN pg_namespace n ON n.oid = i.relnamespace WHERE n.nspname = %s AND i.relname = %s",
(schema, index),
).fetchone()
return row is not None and bool(row[0])
def btree_max_entry_bytes(block_size: int) -> int:
"""`BTMaxItemSize` from nbtree.h for btree version 4: a third of the page left after the page
header, three line pointers and the btree opaque area, rounded down to 8, minus the heap TID a
pivot tuple carries. 2704 on the default 8 KB page, the number in Postgres's error."""
usable: Final = block_size - _PAGE_HEADER_WITH_THREE_LINE_POINTERS_BYTES - _BTREE_PAGE_OPAQUE_BYTES
return (usable // 3) // 8 * 8 - _HEAP_TID_BYTES
def _value_bytes(column: ColumnType, identifier: "sql.Identifier") -> "sql.Composed":
from psycopg import sql
if column.type_name in _OCTET_LENGTH_TYPES:
return sql.SQL("octet_length({})").format(identifier)
return sql.SQL("octet_length({}::text)").format(identifier)
def _aligned(offset: "sql.Composable", align: str) -> "sql.Composed":
from psycopg import sql
unit: Final = _ALIGNMENT.get(align, 8)
return sql.SQL("(({off} + {pad}) / {unit}) * {unit}").format(
off=offset, pad=sql.Literal(unit - 1), unit=sql.Literal(unit)
)
def _value_end(column: ColumnType, identifier: "sql.Identifier", offset: "sql.Composable") -> "sql.Composed":
"""Where this column's value ends when laid down from `offset` the way `index_form_tuple`
does it: NULL takes no room, a fixed-length value is aligned to its type, a variable-length
value of up to 126 bytes gets a 1-byte header and no alignment, a longer one a 4-byte header."""
from psycopg import sql
length: Final = _value_bytes(column, identifier)
placed: Final = (
sql.SQL("{} + {}").format(_aligned(offset, column.align), sql.Literal(column.length))
if not column.is_variable_length
else sql.SQL("CASE WHEN {len} <= {short} THEN {off} + 1 + {len} ELSE {aligned} + {hdr} + {len} END").format(
len=length,
short=sql.Literal(SHORT_VARLENA_MAX_BYTES),
off=offset,
aligned=_aligned(offset, column.align),
hdr=sql.Literal(VARLENA_HEADER_BYTES),
)
)
return sql.SQL("CASE WHEN {col} IS NULL THEN {off} ELSE {placed} END").format(
col=identifier, off=offset, placed=placed
)
def _entry_size_query(check: Check, schema: str, types: tuple[ColumnType, ...], limit_bytes: int) -> "sql.Composed":
"""count(*) and the largest entry over the rows whose btree entry would not fit, plus the
longest value of each variable-length column: an 8-byte tuple header, 8 more for the null
bitmap when a column is NULL, each value laid down in index order, the total rounded up to 8."""
from psycopg import sql
identifiers: Final = tuple(sql.Identifier(column) for column in check.columns)
any_null: Final = sql.SQL(" OR ").join(sql.SQL("{} IS NULL").format(identifier) for identifier in identifiers)
header: Final = sql.SQL("{} + CASE WHEN {} THEN {} ELSE 0 END").format(
sql.Literal(INDEX_TUPLE_HEADER_BYTES), any_null, sql.Literal(NULL_BITMAP_BYTES)
)
innermost: Final = sql.SQL("SELECT {cols}, {hdr} AS off0 FROM {t}").format(
cols=sql.SQL(", ").join(identifiers), hdr=header, t=sql.Identifier(schema, check.table)
)
def lay_down(inner: "sql.Composed", step: tuple[int, tuple[ColumnType, "sql.Identifier"]]) -> "sql.Composed":
position, (column, identifier) = step
return sql.SQL("SELECT *, {end} AS {next} FROM ({inner}) AS {alias}").format(
end=_value_end(column, identifier, sql.Identifier(f"off{position}")),
next=sql.Identifier(f"off{position + 1}"),
inner=inner,
alias=sql.Identifier(f"s{position}"),
)
laid_out: Final = reduce(lay_down, enumerate(zip(types, identifiers, strict=True)), innermost)
entry: Final = _aligned(sql.Identifier(f"off{len(types)}"), "d")
longest: Final = sql.SQL("").join(
sql.SQL(", max({})").format(_value_bytes(column, identifier))
for column, identifier in zip(types, identifiers, strict=True)
if column.is_variable_length
)
return sql.SQL("SELECT count(*), max({entry}){longest} FROM ({rows}) AS entries WHERE {entry} > {limit}").format(
entry=entry, longest=longest, rows=laid_out, limit=sql.Literal(limit_bytes)
)
def _count_query(check: Check, schema: str, types: tuple[ColumnType, ...], limit_bytes: int) -> "sql.Composed":
from psycopg import sql
table: Final = sql.Identifier(schema, check.table)
columns: Final = sql.SQL(", ").join(sql.Identifier(column) for column in check.columns)
match check.kind:
case "too_long":
return _entry_size_query(check, schema, types, limit_bytes)
case "duplicates":
not_null: Final = sql.SQL(" AND ").join(
sql.SQL("{} IS NOT NULL").format(sql.Identifier(column)) for column in check.columns
)
return sql.SQL(
"SELECT count(*) FROM (SELECT 1 FROM {t} WHERE {nn} GROUP BY {cols} HAVING count(*) > 1) AS d"
).format(t=table, nn=not_null, cols=columns)
case "nulls":
return sql.SQL("SELECT count(*) FROM {t} WHERE {col} IS NULL").format(
t=table, col=sql.Identifier(check.columns[0])
)
case "any_row":
return sql.SQL("SELECT count(*) FROM {t}").format(t=table)
case "orphans":
pairs: Final = sql.SQL(" AND ").join(
sql.SQL("parent.{} = child.{}").format(sql.Identifier(theirs), sql.Identifier(own))
for own, theirs in zip(check.columns, check.referenced_columns, strict=True)
)
present: Final = sql.SQL(" AND ").join(
sql.SQL("child.{} IS NOT NULL").format(sql.Identifier(column)) for column in check.columns
)
return sql.SQL(
"SELECT count(*) FROM {t} AS child WHERE {present} AND NOT EXISTS (SELECT 1 FROM {ref} AS parent WHERE {pairs})"
).format(t=table, present=present, ref=sql.Identifier(schema, check.referenced_table or ""), pairs=pairs)
case "check_violations":
return sql.SQL("SELECT count(*) FROM {t} WHERE NOT ({expr})").format(
t=table,
expr=sql.SQL(check.expression or "true"), # pyright: ignore[reportArgumentType] # copied from our own migration file
)
def _applies(
connection: "psycopg.Connection[tuple[object, ...]]",
schema: str,
check: Check,
types: tuple[ColumnType, ...] | None,
) -> bool:
"""False when the check has nothing to count yet: the table or a column is not there
(a pending migration adds it), the index is already valid, or no column can overflow."""
if not _table_exists(connection, schema, check.table):
return False
if check.index is not None and _index_is_valid(connection, schema, check.index):
return False
if check.kind == "any_row":
return types is None
if types is None:
return False
if check.kind == "too_long":
return any(column.can_overflow_btree for column in types)
if check.kind == "orphans":
return check.referenced_table is not None and _table_exists(connection, schema, check.referenced_table)
return True
def _too_long_detail(check: Check, types: tuple[ColumnType, ...], values: tuple[object, ...], limit_bytes: int) -> str:
variable: Final = tuple(
column for column, column_type in zip(check.columns, types, strict=True) if column_type.is_variable_length
)
longest: Final = ", ".join(
f"{column} up to {length} bytes"
for column, length in zip(variable, values[2:], strict=True)
if length is not None
)
return f"with an index entry over {limit_bytes} bytes, largest {values[1]} bytes ({longest})"
def _run_check(
connection: "psycopg.Connection[tuple[object, ...]]",
schema: str,
check: Check,
types: tuple[ColumnType, ...],
limit_bytes: int,
timeout_seconds: float,
) -> Finding | None:
import psycopg
try:
row: Final = connection.execute(_count_query(check, schema, types, limit_bytes)).fetchone()
except psycopg.errors.QueryCanceled:
return CouldNotVerify(check, timeout_seconds)
values: Final[tuple[object, ...]] = row if row is not None else ()
rows: Final = _int(values[0], 0) if values else 0
if rows == 0:
return None
match check.kind:
case "too_long":
return BlockedChange(check, rows, _too_long_detail(check, types, values, limit_bytes))
case "duplicates":
return BlockedChange(check, rows, "duplicated value groups")
case "nulls":
return BlockedChange(check, rows, "NULL")
case "any_row":
return BlockedChange(
check, rows, "rows, and a NOT NULL column with no DEFAULT cannot be added to a table with rows"
)
case "orphans":
return BlockedChange(check, rows, f'with no matching row in "{check.referenced_table}"')
case "check_violations":
return BlockedChange(check, rows, f"failing CHECK ({check.expression})")
def find_blockers(
connection: "psycopg.Connection[tuple[object, ...]]",
schema: str,
checks: tuple[Check, ...],
timeout_seconds: float = DEFAULT_MIGRATION_DATA_CHECK_TIMEOUT_SECONDS,
) -> tuple[Finding, ...]:
"""Run every applicable check read-only, each under `timeout_seconds`. Never writes
and never takes a table lock; a check that times out is a finding, not a pass."""
from psycopg import sql
previous_timeout: Final = connection.execute("SHOW statement_timeout").fetchone()
previous_read_only: Final = connection.execute("SHOW default_transaction_read_only").fetchone()
block_size: Final = connection.execute("SHOW block_size").fetchone()
limit_bytes: Final = btree_max_entry_bytes(int(str(block_size[0])) if block_size else DEFAULT_BLOCK_SIZE)
connection.execute("SET default_transaction_read_only = on")
connection.execute(sql.SQL("SET statement_timeout = {}").format(sql.Literal(int(timeout_seconds * 1000))))
column_type: Final = column_type_lookup(connection, schema)
typed: Final = tuple((check, _column_types(check, column_type)) for check in checks)
applicable: Final = tuple(
(check, types if types is not None else ())
for check, types in typed
if _applies(connection, schema, check, types)
)
results: Final = tuple(
_run_check(connection, schema, check, types, limit_bytes, timeout_seconds) for check, types in applicable
)
connection.execute(
sql.SQL("SET statement_timeout = {}").format(sql.Literal(str(previous_timeout[0]) if previous_timeout else "0"))
)
connection.execute(
sql.SQL("SET default_transaction_read_only = {}").format(
sql.Literal(str(previous_read_only[0]) if previous_read_only else "off")
)
)
return tuple(result for result in results if result is not None)
def _remedy(check: Check) -> str:
joined: Final = ", ".join(check.columns)
match check.kind:
case "too_long":
return f'Shorten or hash the long "{check.table}" ({joined}) values so every row\'s index entry fits.'
case "duplicates":
return f'Remove or merge the duplicate "{check.table}" ({joined}) values.'
case "nulls":
return f'Fill the NULL "{check.table}"."{joined}" values, or give the column a default first.'
case "any_row":
return "Give the new column a DEFAULT in the migration, or add it nullable and fill it afterwards."
case "orphans":
return f'Delete or repoint the "{check.table}" ({joined}) values with no "{check.referenced_table}" row.'
case "check_violations":
return f'Fix the "{check.table}" rows that fail the expression before adding the constraint.'
def _describe(finding: Finding) -> str:
check: Final = finding.check
where: Final = f'"{check.table}"' + (f".{', '.join(check.columns)}" if check.columns else "")
match finding:
case CouldNotVerify(timeout_seconds=seconds):
return (
f" {where}: could not be checked within {seconds:g} s, refusing rather than guessing\n"
f" blocks: {check.statement} ({check.source})\n"
f" raise {MIGRATION_DATA_CHECK_TIMEOUT_ENV_VAR} to give the scan more time"
)
case BlockedChange(rows=rows, detail=detail):
return (
f" {where}: {rows} row(s) {detail}\n"
f" blocks: {check.statement} ({check.source})\n"
f" fix: {_remedy(check).replace(chr(10), chr(10) + ' ' * 8)}"
)
def report(findings: tuple[Finding, ...]) -> str:
"""The operator-facing refusal: one block per finding, then what to do next."""
lines: Final = "\n".join(_describe(finding) for finding in findings)
return (
f"{REFUSAL}\n{lines}\n"
"No schema change was made. Fix the rows and start again, or run `litellm --check_migrations_against_existing_data` with the new image to check again."
)
def pending_change_checks(
connection: "psycopg.Connection[tuple[object, ...]]",
schema: str,
migrations_dir: Path | None,
indexes: tuple[tuple[str, str, tuple[str, ...]], ...],
) -> tuple[Check, ...]:
"""Every check the pending migrations and the request-log indexes need. `migrations_dir`
is None when the schema is pushed rather than migrated, so only the indexes are checked."""
migration_checks: Final = (
tuple(
check
for name, migration_sql in pending_migrations(connection, schema, migrations_dir)
for check in checks_for_migration(name, migration_sql)
) # comprehension-ok: flattening the per-migration checks
if migrations_dir is not None
else ()
)
index_checks: Final = tuple(
check for table, name, columns in indexes for check in checks_for_index(table, name, columns)
) # comprehension-ok: flattening the per-index checks
return migration_checks + index_checks

View file

@ -1,4 +1,5 @@
import random
import re
import time
from collections.abc import Generator, Mapping
from contextlib import contextmanager
@ -7,6 +8,7 @@ from typing import TYPE_CHECKING, Final
from urllib.parse import parse_qsl, urlencode, urlsplit, urlunsplit
from litellm_proxy_extras._logging import logger
from litellm_proxy_extras._redaction import _redact_credentials
from litellm_proxy_extras.prisma_toolchain import (
MIGRATION_LOCK_TIMEOUT_ENV_VAR,
migration_lock_timeout,
@ -14,6 +16,26 @@ from litellm_proxy_extras.prisma_toolchain import (
MIGRATION_LOCK_KEY: Final = int.from_bytes(b"llm_mig2", "big")
PRISMA_ADVISORY_LOCK_KEY: Final = 72707369
_SERVER_ANSWERED_MARKERS: Final = ("FATAL:", "SSL", "certificate")
@dataclass(frozen=True, slots=True)
class AdvisoryLock:
key: int
name: str
holder: str
COORDINATOR_LOCK: Final = AdvisoryLock(
MIGRATION_LOCK_KEY,
"v2 migration coordinator lock",
"another v2 migration resolver",
)
PRISMA_LOCK: Final = AdvisoryLock(
PRISMA_ADVISORY_LOCK_KEY,
"Prisma migration lock",
"another Prisma migration",
)
if TYPE_CHECKING:
import psycopg
@ -49,6 +71,28 @@ def _try_lock(connection: "psycopg.Connection[tuple[object, ...]]", key: int = M
return row is not None and row.acquired
def _coordination_failure(database_url: str, exc: "psycopg.Error", *, connecting: bool) -> str:
import psycopg
prefix: Final = "Lost or could not establish v2 migration coordination with the database: "
raw: Final = str(exc)
unquoted: Final = re.sub(r"\"[^\"]*\"|'[^']*'", "", raw)
detail: Final = _redact_credentials(raw).strip()
if (
not connecting
or not isinstance(exc, psycopg.OperationalError)
or any(marker in unquoted for marker in _SERVER_ANSWERED_MARKERS)
):
return prefix + detail
parsed_url: Final = urlsplit(database_url)
try:
port: Final = parsed_url.port
except ValueError:
return prefix + f"P1001: Can't reach database server. {detail}"
host: Final = parsed_url.hostname or "localhost"
return prefix + f"P1001: Can't reach database server at `{host}:{port or 5432}`. {detail}"
@dataclass(frozen=True, slots=True)
class MigrationCoordinator:
connection: "psycopg.Connection[tuple[object, ...]]"
@ -75,39 +119,45 @@ class MigrationCoordinator:
@contextmanager
def migration_lock(database_url: str) -> Generator[MigrationCoordinator, None, None]:
def migration_lock(
database_url: str,
lock: AdvisoryLock = COORDINATOR_LOCK,
) -> Generator[MigrationCoordinator, None, None]:
import psycopg
wait_seconds: Final = migration_lock_timeout()
deadline: Final = time.monotonic() + wait_seconds
try:
with psycopg.connect(database_url, connect_timeout=10, autocommit=True) as connection:
connection: Final = psycopg.connect(database_url, connect_timeout=10, autocommit=True)
except psycopg.Error as exc:
raise RuntimeError(_coordination_failure(database_url, exc, connecting=True)) from None
try:
with connection:
coordinator: Final = MigrationCoordinator(connection)
logger.info("Waiting for the v2 migration coordinator lock (up to %ss)", wait_seconds)
logger.info("Waiting for the %s (up to %ss)", lock.name, wait_seconds)
while time.monotonic() < deadline:
with connection.transaction():
if _try_lock(connection):
logger.info("Acquired the v2 migration coordinator lock")
if _try_lock(connection, lock.key):
logger.info("Acquired the %s", lock.name)
yield coordinator
coordinator.check_connection()
return
time.sleep(min(random.uniform(0.5, 1.5), max(0.0, deadline - time.monotonic())))
except psycopg.Error as exc:
raise RuntimeError(f"Lost or could not establish v2 migration coordination with the database: {exc}") from exc
raise RuntimeError(_coordination_failure(database_url, exc, connecting=False)) from None
raise RuntimeError(
f"Timed out waiting for another v2 migration resolver after {wait_seconds}s. "
f"Timed out waiting for {lock.holder} after {wait_seconds}s. "
f"Check the running migration or increase {MIGRATION_LOCK_TIMEOUT_ENV_VAR}."
)
@contextmanager
def prisma_migration_lock(database_url: str) -> Generator[None, None, None]:
"""Hold the coordinator lock and Prisma's own lock key as transaction-level locks on one
pinned connection while a Prisma command runs with its session advisory lock disabled, so a
transaction pooler can never strand that lock on an idle backend after the command exits."""
with migration_lock(database_url) as coordinator:
coordinator.acquire_prisma_lock()
"""Hold only Prisma's key as an xact lock on a pinned connection with Prisma's session lock disabled.
This keeps deploy independent of the coordinator key held by request-log index builds."""
with migration_lock(database_url, PRISMA_LOCK) as coordinator:
coordinator.release_statement_snapshot()
yield

View file

@ -0,0 +1,4 @@
SET LOCAL lock_timeout = '10s';
-- AlterTable
ALTER TABLE "LiteLLM_ManagedObjectTable" ADD COLUMN IF NOT EXISTS "end_user_id" TEXT;

View file

@ -0,0 +1,23 @@
SET LOCAL lock_timeout = '10s';
CREATE SEQUENCE IF NOT EXISTS "litellm_environment_variables_revision_seq";
GRANT USAGE ON SEQUENCE "litellm_environment_variables_revision_seq" TO PUBLIC;
CREATE OR REPLACE FUNCTION litellm_bump_environment_variables_revision() RETURNS TRIGGER AS $$
DECLARE
next_revision BIGINT := nextval(format('%I.litellm_environment_variables_revision_seq', TG_TABLE_SCHEMA)::regclass);
BEGIN
IF TG_OP = 'UPDATE' THEN
NEW."reload_revision" := GREATEST(next_revision, OLD."reload_revision" + 1);
ELSE
NEW."reload_revision" := next_revision;
END IF;
RETURN NEW;
END;
$$ LANGUAGE plpgsql;
DROP TRIGGER IF EXISTS "litellm_environment_variables_revision" ON "LiteLLM_Config";
CREATE TRIGGER "litellm_environment_variables_revision"
BEFORE INSERT OR UPDATE ON "LiteLLM_Config"
FOR EACH ROW WHEN (NEW."param_name" = 'environment_variables')
EXECUTE PROCEDURE litellm_bump_environment_variables_revision();

View file

@ -0,0 +1,2 @@
CREATE INDEX CONCURRENTLY IF NOT EXISTS "idx_background_interaction_settlement_settled_at"
ON "LiteLLM_BackgroundInteractionSettlement" ("settled_at");

View file

@ -18,6 +18,12 @@ from dataclasses import dataclass
from typing import TYPE_CHECKING, Final
from litellm_proxy_extras._logging import logger
from litellm_proxy_extras.migration_data_checks import (
checks_for_index,
find_blockers,
migration_data_check_timeout_seconds,
report,
)
from litellm_proxy_extras.migration_lock import held_migration_lock
if TYPE_CHECKING:
@ -355,6 +361,12 @@ def _build_leaf_index(
connection.execute(sql.SQL("DROP INDEX CONCURRENTLY {}").format(sql.Identifier(schema, name)))
elif _adopt_equivalent_index(connection, schema, table, name, index):
return True
blockers: Final = find_blockers(
connection, schema, checks_for_index(table, name, index.columns), migration_data_check_timeout_seconds()
)
if blockers:
logger.warning("Not building index %s on %s. %s", name, table, report(blockers))
return False
logger.info("Building index %s on %s concurrently", name, table)
prefix: Final = sql.SQL("CREATE INDEX CONCURRENTLY IF NOT EXISTS {} ON {} ").format(
sql.Identifier(name), sql.Identifier(schema, table)

View file

@ -1179,6 +1179,7 @@ model LiteLLM_ManagedObjectTable { // for batches or finetuning jobs which use t
org_id String? // creating key's organization at submission time; CheckBatchCost bills org spend against it
api_key String?
request_tags Json? @default("[]")
end_user_id String?
updated_at DateTime @updatedAt
updated_by String?
@ -1980,6 +1981,7 @@ model LiteLLM_BackgroundInteractionSettlement {
outcome String?
@@index([claimed_at], map: "idx_background_interaction_settlement_claimed_at")
@@index([settled_at], map: "idx_background_interaction_settlement_settled_at")
}
model LiteLLM_Lens {

View file

@ -1,4 +1,3 @@
import functools
import glob
import os
import random
@ -13,10 +12,17 @@ from contextlib import AbstractContextManager, nullcontext
from dataclasses import dataclass, replace
from pathlib import Path
from typing import TYPE_CHECKING, Final, Optional, Protocol, Union
from urllib.parse import unquote, urlsplit
from litellm_proxy_extras import prisma_toolchain
from litellm_proxy_extras._logging import logger
from litellm_proxy_extras._redaction import _redact_credentials
from litellm_proxy_extras.migration_data_checks import (
Finding,
find_blockers,
migration_data_check_timeout_seconds,
pending_change_checks,
report,
)
from litellm_proxy_extras.migration_lock import held_migration_lock
from litellm_proxy_extras.prisma_toolchain import (
PRISMA_COMMAND_TIMEOUT_ENV_VAR,
@ -29,7 +35,11 @@ from litellm_proxy_extras.replica_identity import (
REPLICA_IDENTITY_FULL_ENV_VAR,
apply_replica_identity_full,
)
from litellm_proxy_extras.request_log_indexes import ensure_request_log_indexes, filter_request_log_index_diff
from litellm_proxy_extras.request_log_indexes import (
REQUEST_LOG_INDEXES,
ensure_request_log_indexes,
filter_request_log_index_diff,
)
if TYPE_CHECKING:
import psycopg
@ -54,9 +64,7 @@ def _get_prisma_env() -> dict:
if str_to_bool(os.getenv("PRISMA_OFFLINE_MODE")):
# These env vars prevent Prisma from attempting downloads
prisma_env["NPM_CONFIG_PREFER_OFFLINE"] = "true"
prisma_env["NPM_CONFIG_CACHE"] = os.getenv(
"NPM_CONFIG_CACHE", "/app/.cache/npm"
)
prisma_env["NPM_CONFIG_CACHE"] = os.getenv("NPM_CONFIG_CACHE", "/app/.cache/npm")
return prisma_env
@ -87,6 +95,7 @@ class _InvalidIndex:
name: str
table_size: str
MAX_MIGRATE_DEPLOY_ATTEMPTS = 4
LIBPQ_URL_PARAMS: Final = frozenset(
{
@ -136,9 +145,7 @@ class _MigrateAttemptBudget:
_SPEND_LOGS_ALTER_RE = re.compile(r'^ALTER\s+TABLE\s+"LiteLLM_SpendLogs"\s', re.IGNORECASE)
_SPEND_LOGS_ARTIFACT_DROP_RE = re.compile(
r'^DROP\s+TABLE\s+"LiteLLM_SpendLogs_[^"]*"', re.IGNORECASE
)
_SPEND_LOGS_ARTIFACT_DROP_RE = re.compile(r'^DROP\s+TABLE\s+"LiteLLM_SpendLogs_[^"]*"', re.IGNORECASE)
_SPEND_LOGS_PK_CLAUSE_RE = re.compile(
r'^(?:DROP\s+CONSTRAINT\s+"[^"]*_pkey"'
r'|ADD\s+(?:CONSTRAINT\s+"[^"]*"\s+)?PRIMARY\s+KEY\s*\([^)]*\))$',
@ -147,9 +154,9 @@ _SPEND_LOGS_PK_CLAUSE_RE = re.compile(
PARTITIONED_SPEND_LOGS_PUSH_ERROR = (
"LiteLLM_SpendLogs is a partitioned table (see db_scripts/partition_spend_logs.sql), "
"so its primary key must include the partition key (\"startTime\"). `prisma db push` "
'so its primary key must include the partition key ("startTime"). `prisma db push` '
"reconciles the database against schema.prisma, which declares the unpartitioned "
"primary key (\"request_id\"), and Postgres rejects that rewrite with: unique "
'primary key ("request_id"), and Postgres rejects that rewrite with: unique '
"constraint on partitioned table must include all partitioning columns. Start the "
"proxy without --use_prisma_db_push so it uses `prisma migrate deploy`, which only "
"applies shipped migrations and leaves the partitioned primary key alone."
@ -158,9 +165,7 @@ PARTITIONED_SPEND_LOGS_PUSH_ERROR = (
def _without_sql_comments(statement: str) -> str:
return "\n".join(
line
for line in statement.splitlines()
if line.strip() and not line.strip().startswith("--")
line for line in statement.splitlines() if line.strip() and not line.strip().startswith("--")
).strip()
@ -170,7 +175,7 @@ def _without_spend_logs_pk_clauses(statement: str) -> Optional[str]:
return statement
kept = tuple(
clause.strip()
for clause in statement[prefix_match.end():].split(",\n")
for clause in statement[prefix_match.end() :].split(",\n")
if not _SPEND_LOGS_PK_CLAUSE_RE.match(clause.strip())
)
if not kept:
@ -212,51 +217,6 @@ def _max_migration_timestamp(names) -> int:
return max(_migration_timestamp(n) for n in names)
_REDACTED: Final = "REDACTED"
_PASSWORD_QUERY_KEYS: Final = frozenset(("password", "sslpassword"))
@functools.cache
def _secret_shape_redactor() -> Callable[[str], str]:
try:
from litellm._logging import redact_secrets
except ImportError:
return lambda text: text
return redact_secrets
def _url_passwords(url: str) -> frozenset[str]:
try:
parts: Final = urlsplit(url)
except ValueError:
return frozenset()
query_pairs: Final = tuple(pair.partition("=") for pair in parts.query.split("&"))
raw_query_passwords: Final = tuple(
value for key, separator, value in query_pairs if separator and key.lower() in _PASSWORD_QUERY_KEYS
)
raw_passwords: Final = ((parts.password,) if parts.password else ()) + raw_query_passwords
return frozenset(password for password in raw_passwords + tuple(map(unquote, raw_passwords)) if password)
def _configured_database_passwords() -> frozenset[str]:
database_url: Final = os.getenv("DATABASE_URL")
direct_url: Final = os.getenv("DIRECT_URL")
database_passwords: Final = _url_passwords(database_url) if database_url else frozenset()
direct_passwords: Final = _url_passwords(direct_url) if direct_url else frozenset()
return database_passwords | direct_passwords
def _redact_credentials(text: str) -> str:
"""Mask configured database passwords before passing the text to LiteLLM redaction."""
passwords: Final = sorted(_configured_database_passwords(), key=len, reverse=True)
alternation: Final = "|".join(re.escape(password) for password in passwords)
password_pattern: Final = (
re.compile(rf"(?<![A-Za-z0-9_])(?:{alternation})(?![A-Za-z0-9_])", re.IGNORECASE) if passwords else None
)
result: Final = password_pattern.sub(_REDACTED, text) if password_pattern is not None else text
return _secret_shape_redactor()(result)
def _redacted_command(command: object) -> Union[str, tuple[str, ...], list[str]]:
if isinstance(command, tuple):
return tuple(_redact_credentials(str(argument)) for argument in command)
@ -289,8 +249,7 @@ def _get_prisma_command() -> str:
# If not found, log warning and fall back
logger.warning(
f"Prisma CLI not found at {default_cli_path}. "
"Falling back to Python wrapper (may attempt downloads)"
f"Prisma CLI not found at {default_cli_path}. Falling back to Python wrapper (may attempt downloads)"
)
# Fall back to the Python wrapper (will work in online mode)
@ -377,9 +336,7 @@ class ProxyExtrasDBManager:
return True
except subprocess.TimeoutExpired:
logger.warning(
"Migration timed out - the database might be under heavy load."
)
logger.warning("Migration timed out - the database might be under heavy load.")
return False
except subprocess.CalledProcessError as e:
logger.warning(
@ -432,10 +389,12 @@ class ProxyExtrasDBManager:
except ImportError:
return None
from psycopg import sql
cleaned_url: Final = ProxyExtrasDBManager._strip_prisma_query_params(database_url)
ledger_table: Final = psycopg.sql.SQL("{}.{}").format(
psycopg.sql.Identifier(ProxyExtrasDBManager._prisma_schema_param(database_url) or "public"),
psycopg.sql.Identifier("_prisma_migrations"),
ledger_table: Final = sql.SQL("{}.{}").format(
sql.Identifier(ProxyExtrasDBManager._prisma_schema_param(database_url) or "public"),
sql.Identifier("_prisma_migrations"),
)
try:
with psycopg.connect(cleaned_url, connect_timeout=10, autocommit=True) as conn:
@ -546,9 +505,7 @@ class ProxyExtrasDBManager:
return filter_partitioned_spend_logs_diff(without_indexes)
@staticmethod
def _resolve_all_migrations(
migrations_dir: str, schema_path: str, mark_all_applied: bool = True
):
def _resolve_all_migrations(migrations_dir: str, schema_path: str, mark_all_applied: bool = True):
"""
1. Compare the current database state to schema.prisma and generate a migration for the diff.
2. Run prisma migrate deploy to apply any pending migrations.
@ -617,9 +574,7 @@ class ProxyExtrasDBManager:
)
logger.info(f"Applied migration: {mig_file.parent.name}")
except subprocess.CalledProcessError as e:
logger.warning(
f"Failed to apply migration {mig_file.parent.name}: {e.stderr}"
)
logger.warning(f"Failed to apply migration {mig_file.parent.name}: {e.stderr}")
except subprocess.TimeoutExpired:
logger.warning(f"Migration {mig_file.parent.name} timed out.")
return
@ -691,9 +646,7 @@ class ProxyExtrasDBManager:
logger.debug(f"Resolved migration: {migration_name}")
except subprocess.CalledProcessError as e:
if "is already recorded as applied in the database." not in e.stderr:
logger.warning(
f"Failed to resolve migration {migration_name}: {e.stderr}"
)
logger.warning(f"Failed to resolve migration {migration_name}: {e.stderr}")
@staticmethod
def raise_if_lens_rename_pending(connect: LensCheckConnect | None = None) -> None:
@ -753,9 +706,7 @@ class ProxyExtrasDBManager:
cleaned_url = ProxyExtrasDBManager._strip_prisma_query_params(database_url)
try:
with psycopg.connect(
cleaned_url, connect_timeout=10, autocommit=True
) as conn:
with psycopg.connect(cleaned_url, connect_timeout=10, autocommit=True) as conn:
row = conn.execute(
"SELECT 1 "
"FROM pg_partitioned_table pt "
@ -763,10 +714,7 @@ class ProxyExtrasDBManager:
"JOIN pg_namespace n ON n.oid = c.relnamespace "
"WHERE c.relname = 'LiteLLM_SpendLogs' "
" AND n.nspname = %s",
(
ProxyExtrasDBManager._prisma_schema_param(database_url)
or "public",
),
(ProxyExtrasDBManager._prisma_schema_param(database_url) or "public",),
).fetchone()
except (psycopg.OperationalError, psycopg.DatabaseError):
return False
@ -789,16 +737,23 @@ class ProxyExtrasDBManager:
(connection_limit, pool_timeout, schema, pgbouncer, sslaccept, ...) and
translate Prisma's TLS params back, since libpq reads ``sslcert`` as a
client certificate where Prisma reads it as the CA."""
from urllib.parse import parse_qsl, quote, urlencode, urlparse, urlunparse
from urllib.parse import parse_qsl, quote, unquote, urlencode, urlparse, urlunparse
parsed: Final = urlparse(url)
netloc: Final = parsed.netloc
userinfo, separator, hostport = netloc.rpartition("@")
user, password_separator, password = userinfo.partition(":")
canonical_userinfo: Final = quote(unquote(user), safe="") + (
password_separator + quote(unquote(password), safe="") if password_separator else ""
)
canonical_netloc: Final = f"{canonical_userinfo}{separator}{hostport}" if separator else netloc
if not parsed.query:
return url
return url if canonical_netloc == netloc else urlunparse(parsed._replace(netloc=canonical_netloc))
pairs: Final = tuple(parse_qsl(parsed.query))
kept: Final = tuple((k, v) for k, v in pairs if k in LIBPQ_URL_PARAMS)
sslaccept: Final = next((v for k, v in pairs if k == "sslaccept"), None)
libpq_pairs: Final = ProxyExtrasDBManager._libpq_tls_params(kept, sslaccept)
return urlunparse(parsed._replace(query=urlencode(libpq_pairs, quote_via=quote)))
return urlunparse(parsed._replace(netloc=canonical_netloc, query=urlencode(libpq_pairs, quote_via=quote)))
@staticmethod
def _libpq_tls_params(
@ -854,9 +809,7 @@ class ProxyExtrasDBManager:
# it, psycopg3's `with conn` calls COMMIT on clean exit — which
# fails after `UndefinedTable` (fresh DB) leaves the transaction
# in an aborted state.
with psycopg.connect(
cleaned_url, connect_timeout=10, autocommit=True
) as conn:
with psycopg.connect(cleaned_url, connect_timeout=10, autocommit=True) as conn:
try:
rows = conn.execute(
"SELECT migration_name FROM _prisma_migrations "
@ -877,9 +830,7 @@ class ProxyExtrasDBManager:
return
head_newest_ts = _max_migration_timestamp(known)
hostile = {
name for name in unknown if _migration_timestamp(name) > head_newest_ts
}
hostile = {name for name in unknown if _migration_timestamp(name) > head_newest_ts}
if not hostile:
return
@ -970,8 +921,7 @@ class ProxyExtrasDBManager:
if not found:
return True
logger.warning(
"Found %d invalid index(es) left by an interrupted CREATE INDEX "
"CONCURRENTLY, rebuilding: %s",
"Found %d invalid index(es) left by an interrupted CREATE INDEX CONCURRENTLY, rebuilding: %s",
len(found),
", ".join(f"{index.name} (table size {index.table_size})" for index in found),
)
@ -1442,8 +1392,7 @@ class ProxyExtrasDBManager:
prisma_env = _get_prisma_env()
except OSError as e:
logger.error(
"Could not resolve the migrations directory for the REPLICA "
"IDENTITY FULL step, skipping it. Error: %s",
"Could not resolve the migrations directory for the REPLICA IDENTITY FULL step, skipping it. Error: %s",
e,
)
return False
@ -1474,18 +1423,57 @@ class ProxyExtrasDBManager:
Returns:
bool: True if setup was successful, False otherwise
"""
ensure_prisma_toolchain(
prisma_command=_get_prisma_command(), prisma_env=_get_prisma_env()
)
migrated = ProxyExtrasDBManager._run_migrations(
use_migrate=use_migrate, use_v2_resolver=use_v2_resolver
)
ensure_prisma_toolchain(prisma_command=_get_prisma_command(), prisma_env=_get_prisma_env())
ProxyExtrasDBManager.raise_if_rows_block_pending_changes(use_migrate=use_migrate)
migrated = ProxyExtrasDBManager._run_migrations(use_migrate=use_migrate, use_v2_resolver=use_v2_resolver)
if not migrated:
return False
ProxyExtrasDBManager.repair_invalid_indexes()
ProxyExtrasDBManager.apply_replica_identity_full_if_requested()
return True
@staticmethod
def migration_data_check_findings(use_migrate: bool = True) -> tuple[Finding, ...]:
"""What in the existing rows would make a pending migration or request-log index
fail, read with count queries only: no write, no table lock. Empty when nothing
blocks, when no database is configured, or when it cannot be reached, so the
migration step that follows reports the connection problem the way it does today."""
import psycopg
database_url: Final = os.environ.get("DATABASE_URL")
if not database_url:
return ()
direct_url: Final = ProxyExtrasDBManager._strip_prisma_query_params(
os.environ.get("DIRECT_URL") or database_url
)
schema: Final = ProxyExtrasDBManager._prisma_schema_param(database_url) or "public"
migrations_dir: Final = Path(ProxyExtrasDBManager._get_prisma_dir()) / "migrations" if use_migrate else None
indexes: Final = tuple((index.table, index.name, index.columns) for index in REQUEST_LOG_INDEXES)
try:
with psycopg.connect(direct_url, connect_timeout=10, autocommit=True) as connection:
checks: Final = pending_change_checks(connection, schema, migrations_dir, indexes)
return find_blockers(connection, schema, checks, migration_data_check_timeout_seconds())
except psycopg.OperationalError as exc:
logger.warning(
"Could not reach the database for the pre-migration row check: %s", _redact_credentials(str(exc))
)
return ()
@staticmethod
def migration_data_check_report(use_migrate: bool = True) -> Optional[str]:
"""The operator-facing refusal for `migration_data_check_findings`, None when nothing blocks."""
findings: Final = ProxyExtrasDBManager.migration_data_check_findings(use_migrate=use_migrate)
return report(findings) if findings else None
@staticmethod
def raise_if_rows_block_pending_changes(use_migrate: bool) -> None:
"""Refuse to run migrations that the existing rows would make fail, before any is
applied. The RuntimeError carries the report and reaches the operator through the
proxy's "Database migration cannot proceed" exit."""
refusal: Final = ProxyExtrasDBManager.migration_data_check_report(use_migrate=use_migrate)
if refusal is not None:
raise RuntimeError(refusal)
@staticmethod
def build_request_log_indexes(build: Callable[[str, str], bool] = ensure_request_log_indexes) -> bool:
"""Build the indexes in `REQUEST_LOG_INDEXES` on the writer, in the schema the
@ -1552,9 +1540,7 @@ class ProxyExtrasDBManager:
# Skip sanity check when deploy reports no pending migrations —
# DB already matches schema, no drift to correct.
if "No pending migrations to apply" in result.stdout:
logger.info(
"No pending migrations — skipping post-migration sanity check"
)
logger.info("No pending migrations — skipping post-migration sanity check")
return True
# Run sanity check to ensure DB matches schema
@ -1569,9 +1555,7 @@ class ProxyExtrasDBManager:
logger.info(f"prisma db error: {stderr}, e: {e.stdout}")
if "P3009" in stderr:
# Extract the failed migration name from the error message
migration_match = re.search(
r"`(\d+_.*)` migration", stderr
)
migration_match = re.search(r"`(\d+_.*)` migration", stderr)
if migration_match:
failed_migration = migration_match.group(1)
if ProxyExtrasDBManager._is_idempotent_error(stderr):
@ -1579,9 +1563,7 @@ class ProxyExtrasDBManager:
f"Migration {failed_migration} failed due to idempotent error (e.g., column already exists), resolving as applied"
)
try:
ProxyExtrasDBManager._roll_back_migration(
failed_migration
)
ProxyExtrasDBManager._roll_back_migration(failed_migration)
except (
subprocess.CalledProcessError,
subprocess.TimeoutExpired,
@ -1591,9 +1573,7 @@ class ProxyExtrasDBManager:
f"It may already be in a rolled-back state."
)
try:
ProxyExtrasDBManager._resolve_specific_migration(
failed_migration
)
ProxyExtrasDBManager._resolve_specific_migration(failed_migration)
logger.info(
f"✅ Migration {failed_migration} resolved, retrying to apply remaining migrations"
)
@ -1601,9 +1581,7 @@ class ProxyExtrasDBManager:
subprocess.CalledProcessError,
subprocess.TimeoutExpired,
) as resolve_err:
logger.warning(
f"Failed to resolve migration {failed_migration}: {resolve_err}"
)
logger.warning(f"Failed to resolve migration {failed_migration}: {resolve_err}")
# Apply any schema drift not covered by the marked-as-applied migration
ProxyExtrasDBManager._resolve_all_migrations(
migrations_dir,
@ -1611,9 +1589,7 @@ class ProxyExtrasDBManager:
mark_all_applied=False,
)
else:
logger.info(
f"Found failed migration: {failed_migration}, marking as rolled back"
)
logger.info(f"Found failed migration: {failed_migration}, marking as rolled back")
# Mark the failed migration as rolled back
prisma_toolchain.run_prisma(
[
@ -1626,23 +1602,14 @@ class ProxyExtrasDBManager:
timeout=prisma_command_timeout(),
env=_get_prisma_env(),
)
logger.info(
f"✅ Migration {failed_migration} marked as rolled back... retrying"
)
elif (
"P3005" in stderr
and "database schema is not empty" in stderr
):
logger.info(f"✅ Migration {failed_migration} marked as rolled back... retrying")
elif "P3005" in stderr and "database schema is not empty" in stderr:
logger.info(
"Database schema is not empty, creating baseline migration. In read-only file system, please set an environment variable `LITELLM_MIGRATION_DIR` to a writable directory to enable migrations. Learn more - https://docs.litellm.ai/docs/proxy/prod#read-only-file-system"
)
ProxyExtrasDBManager._create_baseline_migration(schema_path)
logger.info(
"Baseline migration created, resolving all migrations"
)
ProxyExtrasDBManager._resolve_all_migrations(
migrations_dir, schema_path
)
logger.info("Baseline migration created, resolving all migrations")
ProxyExtrasDBManager._resolve_all_migrations(migrations_dir, schema_path)
logger.info("✅ All migrations resolved.")
return True
elif "P3018" in stderr:
@ -1650,14 +1617,8 @@ class ProxyExtrasDBManager:
if ProxyExtrasDBManager._is_permission_error(stderr):
# Permission errors should NOT be marked as applied
# Extract migration name for logging
migration_match = re.search(
r"Migration name: (\d+_.*)", stderr
)
migration_name = (
migration_match.group(1)
if migration_match
else "unknown"
)
migration_match = re.search(r"Migration name: (\d+_.*)", stderr)
migration_name = migration_match.group(1) if migration_match else "unknown"
logger.error(
f"❌ Migration {migration_name} failed due to insufficient permissions. "
@ -1667,16 +1628,10 @@ class ProxyExtrasDBManager:
# Mark as rolled back and exit with error
if migration_match:
try:
ProxyExtrasDBManager._roll_back_migration(
migration_name
)
logger.info(
f"Migration {migration_name} marked as rolled back"
)
ProxyExtrasDBManager._roll_back_migration(migration_name)
logger.info(f"Migration {migration_name} marked as rolled back")
except Exception as rollback_error:
logger.warning(
f"Failed to mark migration as rolled back: {rollback_error}"
)
logger.warning(f"Failed to mark migration as rolled back: {rollback_error}")
# Re-raise the error to prevent silent failures
raise RuntimeError(
@ -1691,18 +1646,12 @@ class ProxyExtrasDBManager:
"resolving as applied"
)
# Extract the migration name from the error message
migration_match = re.search(
r"Migration name: (\d+_.*)", stderr
)
migration_match = re.search(r"Migration name: (\d+_.*)", stderr)
if migration_match:
migration_name = migration_match.group(1)
try:
logger.info(
f"Rolling back migration {migration_name}"
)
ProxyExtrasDBManager._roll_back_migration(
migration_name
)
logger.info(f"Rolling back migration {migration_name}")
ProxyExtrasDBManager._roll_back_migration(migration_name)
except (
subprocess.CalledProcessError,
subprocess.TimeoutExpired,
@ -1716,9 +1665,7 @@ class ProxyExtrasDBManager:
f"Resolving migration {migration_name} that failed "
f"due to existing schema objects"
)
ProxyExtrasDBManager._resolve_specific_migration(
migration_name
)
ProxyExtrasDBManager._resolve_specific_migration(migration_name)
logger.info(
f"✅ Migration {migration_name} resolved, "
f"retrying to apply remaining migrations"
@ -1727,9 +1674,7 @@ class ProxyExtrasDBManager:
subprocess.CalledProcessError,
subprocess.TimeoutExpired,
) as resolve_err:
logger.warning(
f"Failed to resolve migration {migration_name}: {resolve_err}"
)
logger.warning(f"Failed to resolve migration {migration_name}: {resolve_err}")
# Apply any schema drift not covered by the marked-as-applied migration
ProxyExtrasDBManager._resolve_all_migrations(
migrations_dir,
@ -1771,14 +1716,8 @@ class ProxyExtrasDBManager:
time.sleep(random.randrange(5, 15))
except subprocess.CalledProcessError as e:
attempts_left = 3 - attempt
retry_msg = (
f" Retrying... ({attempts_left} attempts left)"
if attempts_left > 0
else ""
)
stderr_detail: Final = (
f" stderr: {_redact_credentials(str(e.stderr))}" if e.stderr else ""
)
retry_msg = f" Retrying... ({attempts_left} attempts left)" if attempts_left > 0 else ""
stderr_detail: Final = f" stderr: {_redact_credentials(str(e.stderr))}" if e.stderr else ""
logger.error(
f"The process failed to execute. Details: {_redact_command_error(e)}.{stderr_detail}{retry_msg}"
)

View file

@ -1,6 +1,6 @@
[project]
name = "litellm-proxy-extras"
version = "0.4.109"
version = "0.4.111"
description = "Additional files for the LiteLLM Proxy. Reduces the size of the main litellm package."
readme = "README.md"
requires-python = ">=3.9"
@ -30,7 +30,7 @@ required-version = ">=0.10.9"
module-root = ""
[tool.commitizen]
version = "0.4.109"
version = "0.4.111"
version_files = [
"pyproject.toml:^version",
"../pyproject.toml:litellm-proxy-extras==",

View file

@ -15,6 +15,7 @@ Use the derived conversions directly (`<&'static str>::from(x)` / `.into()`, `st
- `crates/<crate>/tests/` is only for tests that drive several modules together end to end, such as a full request through a route. A test of one type or function never goes there
- A helper shared by several inline test modules goes in a `#[cfg(test)] mod test_support;` at the crate root, and only once a second module needs it
- A test for another crate's item belongs in that crate, not in a downstream one
- Tests that call a real provider go in `crates/<crate>/tests/live/<provider>.rs` behind a `tests/live/main.rs`. Each reads the provider's key from the environment through an `Option` fixture and returns early when it is unset, never `#[ignore]`, so `cargo test` stays offline until the key is exported
- Never set `autotests = false` or hand-list `[[test]]` targets. Every file directly under `tests/` is discovered by cargo, and a shared helper goes in `tests/<name>/mod.rs` or `tests/<subject>/support.rs` so it is not picked up as a test crate of its own
## Test fixtures and cases

View file

@ -876,6 +876,15 @@ dependencies = [
"time",
]
[[package]]
name = "backon"
version = "1.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cffb0e931875b666fc4fcb20fee52e9bbd1ef836fd9e9e04ec21555f9f85f7ef"
dependencies = [
"fastrand",
]
[[package]]
name = "base16ct"
version = "0.2.0"
@ -1248,7 +1257,11 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cfc320937d09e6de266b31b9afb480f197d7a861be86be7cb2ea7e5d1bfffc5e"
dependencies = [
"bytes",
"futures-core",
"memchr",
"pin-project-lite",
"tokio",
"tokio-util",
]
[[package]]
@ -3406,6 +3419,7 @@ dependencies = [
name = "litellm-auth-gcp"
version = "0.1.0"
dependencies = [
"base64 0.22.1",
"gcp_auth",
"google-cloud-auth",
"http 1.4.2",
@ -3641,6 +3655,14 @@ dependencies = [
"strum",
]
[[package]]
name = "litellm-circuit-breaker"
version = "0.1.0"
dependencies = [
"proptest",
"rstest",
]
[[package]]
name = "litellm-config"
version = "0.1.0"
@ -3654,6 +3676,8 @@ dependencies = [
"strum",
"tempfile",
"thiserror 2.0.19",
"tracing",
"tracing-subscriber",
]
[[package]]
@ -3694,8 +3718,15 @@ dependencies = [
name = "litellm-db"
version = "0.1.0"
dependencies = [
"chrono",
"macro_rules_attribute",
"rstest",
"serde",
"serde_json",
"sqlx",
"strum",
"thiserror 2.0.19",
"tokio",
]
[[package]]
@ -3735,6 +3766,7 @@ dependencies = [
"futures-util",
"http-body-util",
"litellm-auth-types",
"litellm-circuit-breaker",
"litellm-config",
"litellm-gateway-auth",
"litellm-gateway-inference",
@ -3743,6 +3775,8 @@ dependencies = [
"litellm-http",
"litellm-inference",
"litellm-llms",
"litellm-redis-client",
"litellm-redis-testing",
"litellm-secrets",
"litellm-tracing",
"rstest",
@ -3922,13 +3956,16 @@ version = "0.1.0"
dependencies = [
"bytes",
"futures-util",
"litellm-coroutine",
"litellm-host",
"parking_lot",
"pyo3",
"pyo3-async-runtimes",
"pythonize",
"rstest",
"serde",
"serde_json",
"thiserror 2.0.19",
"tokio",
]
@ -4079,12 +4116,14 @@ dependencies = [
name = "litellm-inference-responses"
version = "0.1.0"
dependencies = [
"base64 0.22.1",
"bytes",
"futures-util",
"litellm-auth",
"litellm-cache",
"litellm-cache-memory",
"litellm-cache-response",
"litellm-core-utils",
"litellm-host",
"litellm-host-native",
"litellm-http",
@ -4269,6 +4308,28 @@ dependencies = [
"thiserror 2.0.19",
]
[[package]]
name = "litellm-redis-client"
version = "0.1.0"
dependencies = [
"futures-util",
"litellm-circuit-breaker",
"litellm-redis-testing",
"redis",
"rstest",
"thiserror 2.0.19",
"tokio",
]
[[package]]
name = "litellm-redis-testing"
version = "0.1.0"
dependencies = [
"testcontainers-modules",
"thiserror 2.0.19",
"tokio",
]
[[package]]
name = "litellm-router"
version = "0.1.0"
@ -5744,22 +5805,34 @@ dependencies = [
[[package]]
name = "redis"
version = "1.7.0"
version = "1.8.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2acbc41a996f7652b2ddd9dfd98cc4ff602cfd742ae35382f07f608405ab50ed"
checksum = "7acd809e80afa754c63f586955699d8990efe1c4b6613183ee5f027a11fbc8b5"
dependencies = [
"arc-swap",
"arcstr",
"async-lock",
"backon",
"bytes",
"cfg-if",
"combine",
"crc16",
"futures-channel",
"futures-util",
"itoa",
"log",
"num-bigint 0.5.1",
"percent-encoding",
"pin-project-lite",
"rand 0.10.2",
"rustls 0.23.42",
"rustls-native-certs",
"ryu",
"sha1_smol",
"socket2 0.6.5",
"tokio",
"tokio-rustls 0.26.4",
"tokio-util",
"url",
"xxhash-rust",
]
@ -8196,18 +8269,18 @@ dependencies = [
[[package]]
name = "webpki-root-certs"
version = "1.0.9"
version = "1.0.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b96554aa2acc8ccdb7e1c9a58a7a68dd5d13bccc69cd124cb09406db612a1c9b"
checksum = "e49d0e65abc00c142e8f9428da991da9f1a250c39be97459901eb12b9b49f62a"
dependencies = [
"rustls-pki-types",
]
[[package]]
name = "webpki-roots"
version = "1.0.9"
version = "1.0.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7dcd9d09a39985f5344844e66b0c530a33843579125f23e21e9f0f220850f22a"
checksum = "f5f280bbb2b287bc8e7de7e3852945a0ff50d1bbcb5ed42b6c7a568c546df171"
dependencies = [
"rustls-pki-types",
]

View file

@ -69,6 +69,9 @@ litellm-token-counter-huggingface = { path = "crates/token-counter-huggingface"
litellm-token-counter-tiktoken = { path = "crates/token-counter-tiktoken" }
litellm-host-python = { path = "crates/host-python" }
litellm-python-compat = { path = "crates/python-compat" }
litellm-circuit-breaker = { path = "crates/circuit-breaker" }
litellm-redis-client = { path = "crates/redis-client" }
litellm-redis-testing = { path = "crates/redis-testing" }
askama = { version = "0.16.1", default-features = false, features = ["derive", "std"] }
tracing = "0.1"
@ -80,6 +83,7 @@ http = "1"
google-cloud-auth = { version = "1.16.0", default-features = false }
jsonwebtoken = { version = "11.1.0", default-features = false }
hyper-util = { version = "0.1.20", default-features = false, features = ["client-proxy"] }
parking_lot = "0.12.5"
proptest = "1.7.0"
pyo3 = "0.29.2"
pyo3-async-runtimes = { version = "0.29.0", features = ["tokio-runtime"] }
@ -87,6 +91,9 @@ rand = "0.8"
macro_rules_attribute = "0.2.3"
schemars = "1"
reqwest = { version = "0.12", default-features = false, features = ["json", "multipart", "rustls-tls", "http2", "stream"] }
redis = "1.8.0"
redis-test = "1.0.4"
testcontainers-modules = "0.15.0"
qdrant-client = { version = "1.19.0", default-features = false }
uuid = { version = "1", features = ["v4"] }
rstest = "0.26.1"
@ -103,6 +110,7 @@ serde_with = { version = "=3.21.0", default-features = false, features = ["std",
sha2 = "0.10"
syn = { version = "2", default-features = false }
sqlx = { version = "0.9.0", default-features = false, features = ["json", "macros", "postgres", "runtime-tokio", "chrono", "tls-rustls-ring-native-roots"] }
chrono = { version = "0.4.45", default-features = false, features = ["std"] }
proc-macro2 = "1"
quote = "1"
subtle = "2"
@ -123,7 +131,7 @@ moka = { version = "0.12.16", features = ["future"] }
strum = { version = "0.28.0", features = ["derive"] }
url = "2.5.8"
percent-encoding = "2.3"
webpki-roots = "1"
webpki-roots = "1.0.10"
time = { version = "0.3.53", features = ["parsing"] }
criterion = "0.8.2"
fancy-regex = "0.19.2"

View file

@ -11,6 +11,7 @@ google-sdk = ["dep:google-cloud-auth", "dep:http"]
[dependencies]
litellm-auth-types.workspace = true
base64.workspace = true
moka.workspace = true
serde.workspace = true
serde_json.workspace = true

View file

@ -1,4 +1,4 @@
use std::{future::Future, path::Path, pin::Pin, sync::Arc};
use std::{future::Future, pin::Pin, sync::Arc};
use gcp_auth::{CustomServiceAccount, TokenProvider};
use litellm_auth_types::{CredentialPlacement, Error, ErrorDetail, http::apply_credential};
@ -8,7 +8,8 @@ use sha2::{Digest, Sha256};
use crate::{
config::{
CredentialSource, VertexConfig, credential_source, get_vertex_ai_project, non_empty_env,
CredentialSource, VertexConfig, credential_source, get_vertex_ai_project,
is_base64_encoded_json, names_a_credentials_file, non_empty_env,
},
constants::{
CLOUD_PLATFORM_SCOPE, GOOGLE_OAUTH_TOKEN_ENDPOINT, VERTEX_AI_API_KEY_ENV,
@ -176,14 +177,7 @@ impl VertexProviderLoader for GcpProviderLoader {
.map_err(auth_acquisition_error)?,
),
CredentialSource::Trusted(configured) => {
let configured = configured.expose();
let service_account = if Path::new(configured).is_file() {
CustomServiceAccount::from_file(configured)
} else {
CustomServiceAccount::from_json(configured)
}
.map_err(auth_acquisition_error)?;
Arc::new(service_account)
Arc::new(load_trusted_service_account(configured.expose())?)
}
CredentialSource::ApplicationCredentials(path) => {
Arc::new(CustomServiceAccount::from_file(path).map_err(auth_acquisition_error)?)
@ -240,6 +234,23 @@ enum CredentialCacheKey {
Adc,
}
const BASE64_CREDENTIALS: &str = "The `vertex_credentials` value is base64-encoded JSON, not the JSON itself. \
Set it to a readable file path, or to the credentials JSON itself.";
fn load_trusted_service_account(configured: &str) -> Result<CustomServiceAccount, Error> {
if is_base64_encoded_json(configured) {
return Err(Error::InvalidConfiguration(ErrorDetail::Message(
BASE64_CREDENTIALS.into(),
)));
}
if names_a_credentials_file(configured) {
CustomServiceAccount::from_file(configured)
} else {
CustomServiceAccount::from_json(configured)
}
.map_err(auth_acquisition_error)
}
fn auth_acquisition_error(error: gcp_auth::Error) -> Error {
Error::CredentialAcquisition(litellm_auth_types::ErrorDetail::failed(
"Vertex AI credentials",
@ -264,6 +275,73 @@ mod tests {
assert_eq!(validate_request_credentials(credentials).is_ok(), accepted);
}
#[rstest::rstest]
#[case::missing_file(
"/nonexistent/vertex-key.json",
"failed to open application credentials file"
)]
#[case::malformed_json(
" {\"type\": \"service_account\",",
"failed to deserialize ApplicationCredentials"
)]
#[case::quoted_json(
"'{\"type\": \"service_account\"}'",
"failed to deserialize ApplicationCredentials"
)]
#[tokio::test]
async fn trusted_credentials_are_read_as_a_file_unless_they_are_json(
#[case] configured: &str,
#[case] expected: &str,
) {
let Err(error) = GcpProviderLoader
.load(CredentialSource::Trusted(SecretValue::new(configured)))
.await
else {
panic!("{configured:?} should not load");
};
assert!(error.to_string().contains(expected), "{error}");
}
#[rstest::rstest]
#[case::padded("eyJ0eXBlIjogInNlcnZpY2VfYWNjb3VudCJ9")]
#[case::quoted_prefixed_unpadded("\"base64:eyJ0eXBlIjogInNlcnZpY2VfYWNjb3VudCJ9\"")]
#[case::url_safe_wrapped("eyJ0eXBlIjogInNlcnZpY2VfYWNjb3VudCIsICJr\nZXkiOiAiPz8_In0=")]
#[tokio::test]
async fn base64_encoded_credentials_are_named_as_base64(#[case] configured: &str) {
assert!(
trusted_load_error(configured)
.await
.contains("is base64-encoded JSON, not the JSON itself")
);
}
async fn trusted_load_error(configured: &str) -> String {
match GcpProviderLoader
.load(CredentialSource::Trusted(SecretValue::new(configured)))
.await
{
Ok(_) => panic!("{configured:?} should not load"),
Err(error) => error.to_string(),
}
}
#[rstest::rstest]
#[tokio::test]
async fn a_relative_key_file_named_like_json_is_read_as_a_file() {
let key = tempfile::Builder::new()
.prefix("{vertex")
.suffix(".json")
.tempfile_in(".")
.unwrap();
std::fs::write(key.path(), r#"{"project_id":"file-project"}"#).unwrap();
let relative_name = key.path().file_name().unwrap().to_str().unwrap();
let absolute_path = std::fs::canonicalize(key.path()).unwrap();
assert_eq!(
trusted_load_error(relative_name).await,
trusted_load_error(absolute_path.to_str().unwrap()).await
);
}
#[rstest::rstest]
fn inline_and_trusted_credentials_never_share_a_cache_entry() {
assert_ne!(

View file

@ -1,5 +1,9 @@
use std::{collections::BTreeMap, path::Path};
use base64::{
Engine, alphabet,
engine::{DecodePaddingMode, GeneralPurpose, GeneralPurposeConfig},
};
use litellm_auth_types::{Error, ErrorDetail, InputSource, SecretValue, Sourced};
use serde_json::{Map, Value};
@ -80,9 +84,10 @@ pub fn get_vertex_ai_project_from_credentials(
CredentialSource::Inline(configured) => configured.expose().to_string(),
CredentialSource::Trusted(configured) => {
let configured = configured.expose();
match std::fs::read_to_string(configured) {
Ok(contents) if Path::new(configured).is_file() => contents,
_ => configured.to_string(),
if names_a_credentials_file(configured) {
std::fs::read_to_string(configured).ok()?
} else {
configured.to_string()
}
}
CredentialSource::ApplicationCredentials(path) => std::fs::read_to_string(path).ok()?,
@ -105,6 +110,32 @@ pub fn get_vertex_ai_location(
.or_else(|| VertexParams::LOCATION.resolve(&|_| None, env_lookup))
}
const INLINE_JSON_LEADING_NOISE: [char; 7] = [' ', '\t', '\r', '\n', '\u{feff}', '"', '\''];
fn is_inline_json(configured: &str) -> bool {
configured
.trim_start_matches(INLINE_JSON_LEADING_NOISE)
.starts_with('{')
}
pub(crate) fn names_a_credentials_file(configured: &str) -> bool {
!is_inline_json(configured) || Path::new(configured).is_file()
}
const BASE64_ANY_PADDING: GeneralPurpose = GeneralPurpose::new(
&alphabet::STANDARD,
GeneralPurposeConfig::new().with_decode_padding_mode(DecodePaddingMode::Indifferent),
);
pub(crate) fn is_base64_encoded_json(configured: &str) -> bool {
let compact: String = configured.split_whitespace().collect();
let quoted = compact.trim_matches(['"', '\'']);
let encoded = quoted.strip_prefix("base64:").unwrap_or(quoted);
BASE64_ANY_PADDING
.decode(encoded.replace('-', "+").replace('_', "/"))
.is_ok_and(|decoded| serde_json::from_slice::<Map<String, Value>>(&decoded).is_ok())
}
#[derive(Clone, Debug)]
pub enum CredentialSource {
Inline(SecretValue),

View file

@ -93,3 +93,22 @@ fn project_is_read_from_the_credentials_that_would_authenticate() {
None
);
}
#[rstest]
fn project_is_read_from_a_relative_key_file_named_like_json() {
let key = tempfile::Builder::new()
.prefix("{vertex")
.suffix(".json")
.tempfile_in(".")
.unwrap();
std::fs::write(key.path(), r#"{"project_id":"file-project"}"#).unwrap();
let relative_name = key.path().file_name().unwrap().to_str().unwrap();
let from_file = VertexConfig::from_params(&VertexParams {
vertex_credentials: Some(relative_name.into()),
..VertexParams::default()
});
assert_eq!(
get_vertex_ai_project_from_credentials(&from_file, &|_| None).as_deref(),
Some("file-project")
);
}

View file

@ -8,12 +8,12 @@ repository.workspace = true
[dependencies]
litellm-cache.workspace = true
litellm-cache-redis.workspace = true
redis = { version = "1.7.0", features = ["tls-rustls"] }
redis = { workspace = true, features = ["tls-rustls"] }
sha2.workspace = true
[dev-dependencies]
litellm-cache-testing.workspace = true
redis-test = "1.0.4"
redis-test.workspace = true
rstest.workspace = true
serde_json.workspace = true
tokio.workspace = true

View file

@ -7,12 +7,12 @@ repository.workspace = true
[dependencies]
litellm-cache.workspace = true
redis = { version = "1.7.0", features = ["cluster", "tls-rustls"] }
redis = { workspace = true, features = ["cluster", "tls-rustls"] }
r2d2 = "0.8.10"
tokio.workspace = true
[dev-dependencies]
litellm-cache-testing.workspace = true
redis-test = "1.0.4"
redis-test.workspace = true
rstest.workspace = true
serde_json.workspace = true

View file

@ -17,8 +17,8 @@ litellm-cache-gcs.workspace = true
litellm-http = { workspace = true, features = ["test-support"] }
litellm-cache-memory.workspace = true
litellm-cache-redis.workspace = true
redis = "1.7.0"
redis-test = "1.0.4"
redis.workspace = true
redis-test.workspace = true
rstest.workspace = true
tokio.workspace = true
wiremock.workspace = true

View file

@ -8,13 +8,13 @@ repository.workspace = true
[dependencies]
litellm-cache.workspace = true
litellm-cache-redis.workspace = true
redis = { version = "1.7.0", features = ["tls-rustls"] }
redis = { workspace = true, features = ["tls-rustls"] }
sha2.workspace = true
uuid = { version = "1", features = ["v4"] }
[dev-dependencies]
litellm-cache-testing.workspace = true
redis-test = "1.0.4"
redis-test.workspace = true
rstest.workspace = true
serde_json.workspace = true
tokio.workspace = true

View file

@ -9,7 +9,7 @@ use litellm_host::{
interceptors::{RawResponse, RequestContext, WireRequest},
lifecycle::{FailureOrigin, Timing, epoch_seconds},
};
use litellm_host_python::{HookStep, from_py, missing_state, to_py};
use litellm_host_python::{Error as HostError, HookStep, from_py, to_py};
use pyo3::{
exceptions::{PyBaseException, PyException},
gc::{PyTraverseError, PyVisit},
@ -160,7 +160,7 @@ impl LegacyLogging {
self.response
.as_ref()
.map(|response| HookStep::Ready(response.clone_ref(py)))
.ok_or_else(missing_state)
.ok_or_else(|| HostError::MissingState.into())
}
fn dispatch_success(&self, py: Python<'_>) -> PyResult<()> {
@ -209,7 +209,7 @@ impl LegacyLogging {
fn stream_success(&self, py: Python<'_>, stream: &DeliveredStream) -> PyResult<()> {
let logger = self.logger()?;
let billing = self.stream_billing().ok_or_else(missing_state)?;
let billing = self.stream_billing().ok_or(HostError::MissingState)?;
let billed = Streaming::Success.call(
py,
(
@ -525,7 +525,7 @@ impl LegacyLogging {
return self.stream_failure(py);
}
if origin == FailureOrigin::Call && self.logger.is_some() && self.runs_deployment_hooks() {
let error = self.error.as_ref().ok_or_else(missing_state)?;
let error = self.error.as_ref().ok_or(HostError::MissingState)?;
return Ok(HookStep::Await(
DeploymentHooks::after_failure(py, self.call.kwargs(), error, self.call_type())?,
Self::resume_deployment_failure,
@ -536,7 +536,7 @@ impl LegacyLogging {
pub(crate) fn stream_opened(&mut self, py: Python<'_>, head: &Py<PyAny>) -> PyResult<()> {
if self.stream_billing().is_none() {
return Err(missing_state());
return Err(HostError::MissingState.into());
}
if let Some(key) = &self.cache_key {
head.bind(py).set_item("cache_key", key)?;
@ -551,7 +551,7 @@ impl LegacyLogging {
}
pub(crate) fn stream_chunk(&mut self, py: Python<'_>, chunk: &Py<PyAny>) -> PyResult<()> {
let stream = self.stream.as_mut().ok_or_else(missing_state)?;
let stream = self.stream.as_mut().ok_or(HostError::MissingState)?;
if stream.first_chunk.is_none() {
stream.first_chunk = Some(datetime(py, epoch_seconds())?);
}

View file

@ -235,7 +235,7 @@ python_callbacks! {
"async_log_stream_event",
"async_post_mcp_tool_call_hook",
],
Dispatch::Python("litellm.llms.anthropic.pass_through.messages.handler") => [
Dispatch::Python("litellm.llms.anthropic.messages.native.handler") => [
"async_pre_request_hook",
],
Dispatch::Python("litellm.litellm_core_utils.streaming_handler") => [

View file

@ -0,0 +1,14 @@
- `fsm` is a generic state machine runtime and knows nothing about breakers. `breaker` is a generic circuit breaker and knows nothing about Redis. Deciding which errors count as failures belongs to the consumer
- `Machine::step` stays pure. `&self` is config and time comes in on the event, never read inside the machine. The only clock read is `Admission`'s drop, at the call-site boundary
- Transition rules are one `match (state, event)` per machine, with no macro DSL and no `_` arm that would hide a new state from the compiler
- `Shared` counts a transition, bumps the generation and calls the observer only when the state's enum variant changes, matching Python bumping `generation` only when `_state` changes. Data inside a variant (failure counters, `since`) changes without a transition
- The observer runs under the lock so transitions are seen in generation order. Keep it to logging or a metric
- A disabled breaker is `Option<Breaker>::None` at the consumer, not a flag inside the machine. `Config` comes in already resolved, reading `REDIS_CIRCUIT_BREAKER_*` env is the consumer's job
## Parity with Python `RedisCircuitBreaker` (`litellm/caching/redis_cache.py`)
- The rstest table `matches_the_python_redis_circuit_breaker` was checked case by case against the Python breaker. A behavior change here needs the same check, by driving the Python class with the same steps
- Divergences, each on purpose
- monotonic `Instant` instead of `time.time()`, so a wall clock jump can't open or close the breaker
- a probe `Admission` dropped without an outcome records a hard failure and reopens. Python's probe that dies on `CancelledError` (a `BaseException`) records nothing and leaves the breaker HALF_OPEN for good, and a dropped Rust future is common
- `_failure_count` is dropped, it only fed a log line
- no `enabled` flag, see above

View file

@ -0,0 +1,12 @@
[package]
name = "litellm-circuit-breaker"
version = "0.1.0"
edition.workspace = true
license.workspace = true
repository.workspace = true
[dependencies]
[dev-dependencies]
proptest.workspace = true
rstest.workspace = true

View file

@ -0,0 +1,444 @@
use std::{
sync::Arc,
time::{Duration, Instant},
};
use crate::{Machine, Observer, Shared};
#[derive(Clone, Copy, Debug, PartialEq)]
pub struct Config {
pub failure_threshold: u32,
pub recovery_timeout: Duration,
pub timeout_min_duration: Duration,
}
impl Default for Config {
fn default() -> Self {
Self {
failure_threshold: 5,
recovery_timeout: Duration::from_secs(60),
timeout_min_duration: Duration::from_secs(5),
}
}
}
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub enum Failure {
Hard,
Timeout,
}
#[derive(Clone, Debug, PartialEq, Eq)]
pub enum State {
Closed {
hard: u32,
timeouts: u32,
streak_started: Option<Instant>,
},
Open {
since: Instant,
},
HalfOpen,
}
impl State {
const CLOSED: Self = Self::Closed {
hard: 0,
timeouts: 0,
streak_started: None,
};
}
#[derive(Clone, Copy, Debug)]
pub enum Event {
Admit(Instant),
Succeeded,
Failed(Failure, Instant),
}
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub enum Output {
Admitted,
Probe,
Rejected,
Recorded,
}
impl Machine for Config {
type State = State;
type Event = Event;
type Output = Output;
fn step(&self, state: &State, event: Event) -> (State, Output) {
match (state, event) {
(State::Closed { .. }, Event::Admit(_)) => (state.clone(), Output::Admitted),
(State::Open { since }, Event::Admit(now))
if now.duration_since(*since) > self.recovery_timeout =>
{
(State::HalfOpen, Output::Probe)
}
(State::Open { .. } | State::HalfOpen, Event::Admit(_)) => {
(state.clone(), Output::Rejected)
}
(State::Closed { .. } | State::HalfOpen, Event::Succeeded) => {
(State::CLOSED, Output::Recorded)
}
(State::Open { .. }, Event::Succeeded) => (state.clone(), Output::Recorded),
(State::Closed { hard, .. }, Event::Failed(Failure::Hard, now)) => {
let hard = hard + 1;
let next = match hard >= self.failure_threshold {
true => State::Open { since: now },
false => State::Closed {
hard,
timeouts: 0,
streak_started: None,
},
};
(next, Output::Recorded)
}
(
State::Closed {
hard,
timeouts,
streak_started,
},
Event::Failed(Failure::Timeout, now),
) => {
let timeouts = timeouts + 1;
let started = streak_started.unwrap_or(now);
let opens = timeouts >= self.failure_threshold
&& now.duration_since(started) >= self.timeout_min_duration;
let next = match opens {
true => State::Open { since: now },
false => State::Closed {
hard: *hard,
timeouts,
streak_started: Some(started),
},
};
(next, Output::Recorded)
}
(State::Open { .. } | State::HalfOpen, Event::Failed(_, now)) => {
(State::Open { since: now }, Output::Recorded)
}
}
}
}
#[derive(Clone)]
pub struct Breaker(Arc<Shared<Config>>);
#[derive(Debug, PartialEq, Eq)]
pub struct Rejected;
pub struct Admission {
breaker: Breaker,
generation: u64,
probe: bool,
settled: bool,
}
impl Breaker {
pub fn new(config: Config, observer: Observer<State>) -> Self {
Self(Arc::new(Shared::new(config, State::CLOSED, observer)))
}
pub fn admit(&self, now: Instant) -> Result<Admission, Rejected> {
let (output, generation) = self.0.fire(Event::Admit(now));
let probe = match output {
Output::Admitted => false,
Output::Probe => true,
Output::Rejected | Output::Recorded => return Err(Rejected),
};
Ok(Admission {
breaker: self.clone(),
generation,
probe,
settled: false,
})
}
pub fn failed(&self, failure: Failure, now: Instant) {
self.0.fire(Event::Failed(failure, now));
}
pub fn state(&self) -> State {
self.0.state()
}
}
impl Admission {
pub fn probe(&self) -> bool {
self.probe
}
pub fn succeeded(mut self) {
self.settled = true;
self.breaker.0.fire_if(self.generation, Event::Succeeded);
}
pub fn failed(mut self, failure: Failure, now: Instant) {
self.settled = true;
self.breaker.failed(failure, now);
}
}
impl Drop for Admission {
fn drop(&mut self) {
if self.probe && !self.settled {
self.breaker.0.fire_if(
self.generation,
Event::Failed(Failure::Hard, Instant::now()),
);
}
}
}
#[cfg(test)]
mod tests {
use std::{
collections::HashMap,
sync::{Arc, Mutex},
time::{Duration, Instant},
};
use proptest::prelude::*;
use rstest::rstest;
use super::{Admission, Breaker, Config, Failure, State};
use crate::Transition;
#[derive(Clone, Copy, Debug)]
enum Step {
Fail(f64, Failure),
Admit(f64, bool),
Succeed(usize),
}
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
enum Phase {
Closed,
Open,
HalfOpen,
}
use Phase::{Closed as C, HalfOpen as H, Open as O};
#[rstest]
fn dropping_an_old_probe_does_not_undo_a_newer_recovery() {
let breaker = Breaker::new(
Config {
failure_threshold: 1,
recovery_timeout: Duration::from_secs(1),
..Config::default()
},
Box::new(|_| {}),
);
let now = Instant::now();
let pending = breaker.admit(now).unwrap();
breaker.failed(Failure::Hard, now);
let old_probe = breaker.admit(now + Duration::from_secs(2)).unwrap();
assert!(old_probe.probe());
pending.failed(Failure::Hard, now + Duration::from_secs(2));
let new_probe = breaker.admit(now + Duration::from_secs(4)).unwrap();
assert!(new_probe.probe());
new_probe.succeeded();
assert_eq!(breaker.state(), State::CLOSED);
drop(old_probe);
assert_eq!(breaker.state(), State::CLOSED);
assert!(!breaker.admit(now + Duration::from_secs(4)).unwrap().probe());
}
fn hard(at: f64) -> Step {
Step::Fail(at, Failure::Hard)
}
fn timeout(at: f64) -> Step {
Step::Fail(at, Failure::Timeout)
}
fn five_hard_at_zero() -> Vec<Step> {
vec![hard(0.0); 5]
}
fn phase(state: &State) -> Phase {
match state {
State::Closed { .. } => Phase::Closed,
State::Open { .. } => Phase::Open,
State::HalfOpen => Phase::HalfOpen,
}
}
fn quiet() -> Breaker {
Breaker::new(Config::default(), Box::new(|_| {}))
}
#[rstest]
#[case::five_hard_failures_open(vec![hard(0.0), hard(1.0), hard(2.0), hard(3.0), hard(4.0)], vec![C, C, C, C, O])]
#[case::a_success_resets_the_hard_failure_count(
vec![hard(0.0), hard(1.0), hard(2.0), hard(3.0), Step::Admit(4.0, true), Step::Succeed(4), hard(5.0), hard(6.0), hard(7.0), hard(8.0)],
vec![C; 10]
)]
#[case::a_timeout_streak_opens_only_after_the_min_duration(
vec![timeout(0.0), timeout(1.0), timeout(2.0), timeout(3.0), timeout(4.9), timeout(5.0)],
vec![C, C, C, C, C, O]
)]
#[case::a_hard_failure_restarts_the_timeout_streak(
vec![timeout(0.0), timeout(1.0), timeout(2.0), timeout(3.0), hard(4.0), timeout(5.0), timeout(6.0), timeout(7.0), timeout(8.0), timeout(9.9), timeout(10.0)],
vec![C, C, C, C, C, C, C, C, C, C, O]
)]
#[case::the_probe_waits_until_strictly_past_the_recovery_timeout(
[five_hard_at_zero(), vec![Step::Admit(60.0, false), Step::Admit(60.5, true)]].concat(),
vec![C, C, C, C, O, O, H]
)]
#[case::a_probe_success_closes(
[five_hard_at_zero(), vec![Step::Admit(61.0, true), Step::Succeed(5)]].concat(),
vec![C, C, C, C, O, H, C]
)]
#[case::a_probe_failure_reopens_with_a_fresh_timer(
[five_hard_at_zero(), vec![Step::Admit(61.0, true), hard(62.0), Step::Admit(122.0, false), Step::Admit(122.5, true)]].concat(),
vec![C, C, C, C, O, H, O, O, H]
)]
#[case::only_one_probe_at_a_time(
[five_hard_at_zero(), vec![Step::Admit(61.0, true), Step::Admit(61.5, false)]].concat(),
vec![C, C, C, C, O, H, H]
)]
#[case::a_late_failure_while_open_pushes_back_the_probe(
[five_hard_at_zero(), vec![hard(30.0), Step::Admit(61.0, false), Step::Admit(90.5, true)]].concat(),
vec![C, C, C, C, O, O, O, H]
)]
#[case::a_success_from_before_the_breaker_moved_is_ignored(
vec![Step::Admit(0.0, true), hard(1.0), hard(1.0), hard(1.0), hard(1.0), hard(1.0), Step::Admit(62.0, true), Step::Succeed(0), Step::Succeed(6)],
vec![C, C, C, C, C, O, H, H, C]
)]
fn matches_the_python_redis_circuit_breaker(
#[case] steps: Vec<Step>,
#[case] expected: Vec<Phase>,
) {
let breaker = quiet();
let t0 = Instant::now();
let mut admissions: HashMap<usize, Admission> = HashMap::new();
let phases: Vec<Phase> = steps
.iter()
.enumerate()
.map(|(index, step)| {
match *step {
Step::Fail(at, failure) => {
breaker.failed(failure, t0 + Duration::from_secs_f64(at))
}
Step::Admit(at, admitted) => {
let admission = breaker.admit(t0 + Duration::from_secs_f64(at));
assert_eq!(admission.is_ok(), admitted, "step {index}");
admissions.extend(admission.ok().map(|admission| (index, admission)));
}
Step::Succeed(admitted_at) => {
admissions.remove(&admitted_at).unwrap().succeeded()
}
}
phase(&breaker.state())
})
.collect();
assert_eq!(phases, expected);
}
#[rstest]
fn a_dropped_probe_reopens_instead_of_wedging_half_open() {
let breaker = quiet();
let t0 = Instant::now();
(0..5).for_each(|_| breaker.failed(Failure::Hard, t0));
let probe = breaker.admit(t0 + Duration::from_secs(61)).unwrap();
assert!(probe.probe());
drop(probe);
assert!(matches!(breaker.state(), State::Open { .. }));
}
#[rstest]
fn a_dropped_regular_call_records_nothing() {
let breaker = quiet();
let t0 = Instant::now();
breaker.failed(Failure::Hard, t0);
drop(breaker.admit(t0).unwrap());
assert_eq!(
breaker.state(),
State::Closed {
hard: 1,
timeouts: 0,
streak_started: None
}
);
}
#[derive(Clone, Copy, Debug)]
enum Action {
Admit,
SucceedOldest,
FailOldest(Failure),
Fail(Failure),
}
fn action() -> impl Strategy<Value = (u64, Action)> {
let failure = prop_oneof![Just(Failure::Hard), Just(Failure::Timeout)];
let action = prop_oneof![
Just(Action::Admit),
Just(Action::SucceedOldest),
failure.clone().prop_map(Action::FailOldest),
failure.prop_map(Action::Fail),
];
(0u64..30_000, action)
}
proptest! {
#[test]
fn open_never_closes_directly_and_each_half_open_visit_has_one_probe(
actions in prop::collection::vec(action(), 1..200)
) {
let transitions: Arc<Mutex<Vec<Transition<State>>>> = Arc::default();
let seen = Arc::clone(&transitions);
let config = Config {
failure_threshold: 3,
recovery_timeout: Duration::from_secs(10),
timeout_min_duration: Duration::from_secs(2),
};
let breaker = Breaker::new(config, Box::new(move |t| seen.lock().unwrap().push(t.clone())));
let t0 = Instant::now();
let mut now = t0;
let mut pending: Vec<Admission> = Vec::new();
let mut probes_per_visit: Vec<u32> = Vec::new();
for (advance_ms, action) in actions {
now += Duration::from_millis(advance_ms);
match action {
Action::Admit => {
if let Ok(admission) = breaker.admit(now) {
if admission.probe() {
probes_per_visit.push(1);
}
pending.push(admission);
}
}
Action::SucceedOldest if !pending.is_empty() => pending.remove(0).succeeded(),
Action::FailOldest(failure) if !pending.is_empty() => pending.remove(0).failed(failure, now),
Action::SucceedOldest | Action::FailOldest(_) => {}
Action::Fail(failure) => breaker.failed(failure, now),
}
}
let transitions = transitions.lock().unwrap();
let open_to_closed = transitions
.iter()
.filter(|t| matches!((&t.from, &t.to), (State::Open { .. }, State::Closed { .. })))
.count();
let half_open_visits = transitions.iter().filter(|t| t.to == State::HalfOpen).count();
prop_assert_eq!(open_to_closed, 0);
prop_assert_eq!(probes_per_visit.len(), half_open_visits);
}
}
}

View file

@ -0,0 +1,174 @@
use std::{
fmt::Debug,
mem,
sync::{Mutex, PoisonError},
};
pub trait Machine {
type State: Clone + Debug;
type Event;
type Output;
fn step(&self, state: &Self::State, event: Self::Event) -> (Self::State, Self::Output);
}
#[derive(Clone, Debug)]
pub struct Transition<S> {
pub from: S,
pub to: S,
pub generation: u64,
}
pub type Observer<S> = Box<dyn Fn(&Transition<S>) + Send + Sync>;
pub struct Shared<M: Machine> {
machine: M,
current: Mutex<(M::State, u64)>,
observer: Observer<M::State>,
}
impl<M: Machine> Shared<M> {
pub fn new(machine: M, initial: M::State, observer: Observer<M::State>) -> Self {
Self {
machine,
current: Mutex::new((initial, 0)),
observer,
}
}
pub fn fire(&self, event: M::Event) -> (M::Output, u64) {
let mut current = self.current.lock().unwrap_or_else(PoisonError::into_inner);
self.apply(&mut current, event)
}
pub fn fire_if(&self, generation: u64, event: M::Event) -> Option<M::Output> {
let mut current = self.current.lock().unwrap_or_else(PoisonError::into_inner);
(current.1 == generation).then(|| self.apply(&mut current, event).0)
}
pub fn state(&self) -> M::State {
self.current
.lock()
.unwrap_or_else(PoisonError::into_inner)
.0
.clone()
}
fn apply(&self, current: &mut (M::State, u64), event: M::Event) -> (M::Output, u64) {
let (to, output) = self.machine.step(&current.0, event);
if mem::discriminant(&to) != mem::discriminant(&current.0) {
let from = mem::replace(&mut current.0, to.clone());
current.1 += 1;
(self.observer)(&Transition {
from,
to,
generation: current.1,
});
} else {
current.0 = to;
}
(output, current.1)
}
}
#[cfg(test)]
mod tests {
use std::sync::{Arc, Mutex};
use rstest::rstest;
use super::{Machine, Shared, Transition};
struct Wrapping(u32);
#[derive(Clone, Debug, PartialEq, Eq)]
enum Counter {
Low(u32),
High(u32),
}
enum Tick {
Up,
Stay,
}
impl Machine for Wrapping {
type State = Counter;
type Event = Tick;
type Output = ();
fn step(&self, state: &Counter, event: Tick) -> (Counter, ()) {
let value = match (state, event) {
(Counter::Low(_) | Counter::High(_), Tick::Stay) => return (state.clone(), ()),
(Counter::Low(value) | Counter::High(value), Tick::Up) => (value + 1) % self.0,
};
match value < self.0 / 2 {
true => (Counter::Low(value), ()),
false => (Counter::High(value), ()),
}
}
}
type Seen = Arc<Mutex<Vec<Transition<Counter>>>>;
fn recorded(wrap: u32) -> (Shared<Wrapping>, Seen) {
let transitions: Seen = Arc::default();
let seen = Arc::clone(&transitions);
let shared = Shared::new(
Wrapping(wrap),
Counter::Low(0),
Box::new(move |transition| seen.lock().unwrap().push(transition.clone())),
);
(shared, transitions)
}
#[rstest]
fn concurrent_events_are_all_applied_and_transitions_are_observed_in_generation_order() {
let (shared, transitions) = recorded(4);
std::thread::scope(|scope| {
for _ in 0..64 {
scope.spawn(|| {
for _ in 0..100 {
shared.fire(Tick::Up);
}
});
}
});
assert_eq!(shared.state(), Counter::Low(6400 % 4));
let generations: Vec<u64> = transitions
.lock()
.unwrap()
.iter()
.map(|t| t.generation)
.collect();
assert_eq!(generations, (1..=3200).collect::<Vec<u64>>());
}
#[rstest]
fn a_step_that_keeps_the_variant_neither_counts_nor_notifies() {
let (shared, transitions) = recorded(4);
let (_, after_inner_change) = shared.fire(Tick::Up);
let (_, after_self_loop) = shared.fire(Tick::Stay);
assert_eq!(shared.state(), Counter::Low(1));
assert_eq!((after_inner_change, after_self_loop), (0, 0));
assert!(transitions.lock().unwrap().is_empty());
}
#[rstest]
fn fire_if_skips_an_event_from_an_older_generation() {
let (shared, _) = recorded(4);
shared.fire(Tick::Up);
let (_, generation) = shared.fire(Tick::Up);
let stale = shared.fire_if(generation - 1, Tick::Up);
let current = shared.fire_if(generation, Tick::Up);
assert!(stale.is_none());
assert!(current.is_some());
assert_eq!(shared.state(), Counter::High(3));
}
}

Some files were not shown because too many files have changed in this diff Show more