Apply normalize_mai_image_usage before building ImageResponse so token-based
cost calculation works when Azure returns num_output_tokens fields.
Co-authored-by: Cursor <cursoragent@cursor.com>
Add MAI image edit support with usage normalization for Azure response format,
and register MAI-Image-2.5-Flash and MAI-Image-2e pricing in the model map.
Co-authored-by: Cursor <cursoragent@cursor.com>
Route azure_ai MAI models to /mai/v1/images/generations and map OpenAI size to width/height for the serverless API.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini-realtime): use GA event names for Pipecat 1.3.x compatibility
Pipecat v1.3.0 adopted the OpenAI Realtime API GA event naming:
response.audio.delta -> response.output_audio.delta
response.text.delta -> response.output_text.delta
response.audio.done -> response.output_audio.done
response.text.done -> response.output_text.done
The proxy was still emitting the old beta names; Pipecat's
`parse_server_event` raises "Unimplemented server event type" for any
unknown type, which killed the receive task handler and broke audio
playback and tool-call delivery.
Also:
- conversation.item.created -> conversation.item.added (already handled)
- client audio is buffered until backend setupComplete in deferred mode
- call_id fallback UUID when Gemini returns empty id
- status_details / token detail fields added to Pydantic-strict events
The _GA_TO_BETA_EVENT_TYPES map in RealTimeStreaming already translates
GA names back to beta for clients that opt in with the openai-beta
header, so legacy clients are unaffected.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini-realtime): address greptile review comments
- emit outputTranscription as response.output_audio_transcript.delta
instead of suppressing it; GA_TO_BETA map handles translation for
legacy clients
- cap pre-setup audio buffer at 200 frames to prevent memory exhaustion;
log a warning when the limit is hit and additional frames are dropped
- log remaining dropped message count on flush error
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini-realtime): address veria review comments
- remove unused OpenAIRealtimeConversationItemCreated import
- fix guardrail bypass: semantic_vad early-return now preserves
create_response when set so a guardrail-injected create_response:false
is not silently dropped
- add per-connection 10 MB byte cap alongside the 200-frame count cap
for the pre-setup audio buffer to prevent memory exhaustion
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini-realtime): fix mypy arg-type on _finalize_gemini_live_setup
setup parameter typed as BidiGenerateContentSetup to match the TypedDict
passed at both call sites; was dict which mypy rejected.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini-realtime): widen _finalize_gemini_live_setup to Dict[str, Any]
BidiGenerateContentSetup (TypedDict) is a subtype of Dict[str,Any] so
both call sites (one passing a plain dict, one passing the TypedDict)
satisfy mypy.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini-realtime): cast BidiGenerateContentSetup to Dict at _finalize call site
mypy rejects TypedDict as dict[str, Any] argument; cast at the call site
where follow_up_setup is BidiGenerateContentSetup to satisfy the checker.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Fix Gemini realtime beta compatibility
* Fix deferred Gemini setup audio ordering
* fix: preserve Gemini audio transcript ids
* fix(realtime): cap pre-setup client buffer on all append paths
Route every append to the deferred-setup pending buffer through the
per-connection message/byte caps. Previously only the audio-buffer
fast path enforced the caps; once one frame was buffered, a client
that withheld session.update could stream arbitrary frames into
_pending_messages_until_setup unbounded and exhaust proxy memory.
* style(gemini-realtime): apply black formatting to transformation.py
* fix(gemini-realtime): log beta-translation fallback and name native-audio marker
Surface the previously swallowed exception in _send_event_to_client so a
failed GA->beta translation is observable instead of silently forwarding the
untranslated event. Extract the native-audio model substring used by
_finalize_gemini_live_setup into a named constant documenting why speechConfig
is dropped on those setups.
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(proxy): match passthrough registry routes bare-to-bare with SERVER_ROOT_PATH
After #28547, get_request_route strips the deployment prefix while registry
lookup still re-inflated stored paths via SERVER_ROOT_PATH, causing 404s
under paths like /llmproxy/ml. Compare normalized bare routes in both
is_registered_pass_through_route and get_registered_pass_through_route.
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(proxy): patch utils.get_server_root_path in passthrough auth tests
After removing get_server_root_path from pass_through_endpoints, route
and JWT tests must mock litellm.proxy.utils where normalization reads it.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(gemini): keep googleSearch with server-side tools and googleMaps JSON schema
Wire include_server_side_tool_invocations through completion() so mixed
google_search and function tools are not dropped on Gemini 3+. Rewrite
generationConfig to responseFormat when googleMaps is used with JSON schema.
Fixes#27479Fixes#29451
Co-authored-by: Cursor <cursoragent@cursor.com>
* address greptile review feedback (greploop iteration 1)
* style: fix black formatting in main.py for py312 compat
* Fix Gemini Google Maps extra_body JSON rewrite
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
* ci(ui): add frontend-lint job enforcing prettier and eslint on changed files
Lints only the files a PR adds or modifies under ui/litellm-dashboard,
so new and touched code must be prettier-clean and eslint-clean while the
existing tree is grandfathered. Skips cleanly when a PR touches no
lintable UI files. This lets us adopt the formatters incrementally
without a repo-wide reformat
* ci(ui): write frontend-lint file lists to $RUNNER_TEMP
Keep the prettier/eslint changed-file lists out of the checkout dir so
they cannot collide with a future source file of the same name
* lint(ui): baseline existing eslint findings so only new ones block
Capture the current error-level eslint findings (318 across 183 files)
in a committed suppressions baseline via eslint --suppress-all. Every
rule stays at its error severity, so any newly introduced violation
fails the frontend-lint gate, while the existing tree is grandfathered;
touching a legacy file never forces fixing its pre-existing issues. CI
runs eslint with --pass-on-unpruned-suppressions so that fixing a
baselined issue does not fail on a now-stale suppression, and the
generated baseline is prettier-ignored since eslint owns its format.
Burn the baseline down over time with eslint --prune-suppressions
* lint(ui): enforce a count budget for explicit any
Make @typescript-eslint/no-explicit-any a warning and cap the total
instead of hard-blocking each new one. A frontend-lint step counts the
repo-wide explicit any and fails only when it exceeds the committed
budget in eslint-any-budget.json. max starts at 2031, ten above the
current 2021, so the next ten land as warnings and the build fails once
that headroom is gone. Lower max over time toward target to ratchet the
count down. New anys still surface as warnings on changed files via the
normal eslint step
* lint(ui): enable zero-cost rules no-var, no-self-assign, react/no-danger
These have no existing violations, so they need no baseline; turning them
on purely blocks new instances. react/no-danger guards against new
dangerouslySetInnerHTML (XSS), no-var enforces let/const, and
no-self-assign catches self-assignment typos. no-debugger is already
enforced by the recommended preset
* lint(ui): add baselined complexity rules
Enable complexity:20, max-depth:4, max-params:4, max-nested-callbacks:4,
with thresholds set near the codebase p99 so only genuine outliers are
flagged. The 272 existing over-threshold functions are grandfathered in
the suppressions baseline; new over-threshold functions block. Lower the
thresholds over time to ratchet complexity down. max-lines-per-function
is intentionally left off since React components are legitimately long
* lint(ui): ban new raw fetch, standardize on React Query
Add a no-restricted-syntax rule flagging bare fetch() calls, pointing
contributors at React Query (@tanstack/react-query). The rule is not
exempted anywhere, including the already-bloated networking.tsx, so all
331 existing fetch calls are grandfathered but no new ones can be added
there or elsewhere. New data access goes through React Query, and the
networking layer can be migrated out and pruned from the baseline over
time
* lint(ui): ban new @tremor/react imports
Add a no-restricted-imports rule flagging imports from @tremor/react so
tremor is phased out rather than spread further. The 232 existing tremor
imports are grandfathered in the baseline; new ones block and point at
antd. Migrate components off tremor and prune the baseline over time
* lint(ui): widen explicit-any budget headroom to 2040
Raise max from 2031 to 2040, giving ~19 of slack over the current 2021
instead of 10
* style(ui): prettier-format eslint.config.mjs
The frontend-lint gate flagged its own config file. Format it so the
prettier check on this PR's changed files passes
* lint(ui): soften complexity and max-depth to warnings
These two are smell metrics with arbitrary thresholds where a legit new
function can trip them, so make them advisory rather than hard-blocking.
They drop out of the baseline (now 963). max-params, max-nested-callbacks,
and the react-hooks rules stay strict since those are clear-cut
* lint(ui): move complexity and max-depth to the count-budget pattern
Generalize the explicit-any budget into a shared lint-budget mechanism:
eslint-budgets.json maps a rule to {max, target} and check-lint-budgets.mjs
counts each across the repo and fails when a count exceeds its max.
complexity (129, max 140) and max-depth (61, max 70) now use the same
slack-plus-counter model as explicit-any (2021, max 2040): they warn
per-file and the build only fails if the repo-wide total crosses the
ceiling. Lower each max toward its target over time
* docs(ui): note pruning the eslint suppressions baseline when fixing lint debt
Streaming responses from the proxy (/chat/completions, /v1/messages,
/v1/responses, assistants) all return through create_response() but never
sent the headers that tell an intermediary reverse proxy not to buffer the
SSE stream. nginx with the default proxy_buffering, k8s ingress-nginx, and
Envoy/Istio sidecars therefore hold the whole stream and release it in one
batch, which looks like a broken/buffered stream to the client even though
litellm is yielding chunks incrementally.
Add Cache-Control: no-cache and X-Accel-Buffering: no to every
StreamingResponse create_response() returns, matching what the proxy already
does for its own usage/policy SSE endpoints. Fixes#28384.
#29612 exempts UI/CLI session tokens from the key budget ceiling when they
create a team key, keyed on data.team_id. That value is read after the
default_key_generate_params loop can populate team_id, so on deployments that
set default_key_generate_params.team_id a request the caller did not scope to a
team is treated as a team key and skips the ceiling. Capture _requested_team_id
before defaults run and key the exemption off it, mirroring how
_requested_max_budget is already captured. Requests the caller did not scope to a
team keep the ceiling.
ESLint 9 defaults to flat config and eslint-config-next was pinned at 15
while Next is on 16, so eslint only ran with ESLINT_USE_FLAT_CONFIG=false
and next lint is gone on Next 16. Replace .eslintrc.json with a native
flat eslint.config.mjs (config-next 16 ships flat configs, so no
FlatCompat shim is needed), bump eslint-config-next to 16.2.6, add
@eslint/js and typescript-eslint as explicit devDeps for the recommended
rule sets, and point the lint script at eslint directly.
This only makes eslint runnable on modern tooling; it does not wire it
into CI. The same rules carry over (next/core-web-vitals, eslint and
typescript-eslint recommended, prettier, unused-imports)
Allow realtime event transcript fields to be nullable so GA conversation.item payloads with transcript=null don't fail logging normalization and suppress success callbacks.
Co-authored-by: Cursor <cursoragent@cursor.com>
Non-admin users creating a team key through the UI were rejected with
"max_budget cannot exceed the caller's own max_budget (0.25)". The request is
authenticated by a UI/CLI session token whose max_budget is the per-session chat
spend cap (max_ui_session_budget, default $0.25), and the delegated-authority
budget ceiling (GHSA-q775-qw9r-2r4g) treated that cap as a delegation limit.
Skip the ceiling only when a session token creates a team key (data.team_id set);
that key's spend is bounded by the team budget at request time. Personal keys and
every other non-admin caller keep the ceiling, so a session token cannot mint an
arbitrary-budget personal key.
* Fix remaining VCR live-call leaks
* test(vcr): dedupe live-test helpers and drop spurious kwargs
Extract the duplicated isVertexQuotaError/runVertexRequestOrSkip Vertex
quota-skip helpers into tests/pass_through_tests/vertex_test_helpers.js and the
duplicated _skip_live_prompt_caching_test guard into tests/_live_test_helpers.py
so each lives in one place. In test_aarun_thread_litellm, build a separate
message_data carrying role/content for add_message and a thread_data without
them for run_thread/run_thread_stream/get_messages, which no longer receive the
spurious message fields.
* test(overhead): assert mock transport is exercised in non-streaming and stream tests
* fix(ci): keep coverage rename green when a parallel node runs no tests
local_testing_part1 and local_testing_part2 run with parallelism 4. When
CircleCI reruns only the failed tests, the failed test lands on a single
node and the other nodes receive an empty bucket, so pytest never writes
coverage.xml or .coverage. The unguarded "mv coverage.xml ..." then exits
1 and turns the whole job red even though the rerun passed; the next
persist_to_workspace step would fail the same way on the missing paths.
Guard the rename so a node with no coverage emits empty placeholders
instead. coverage combine tolerates the empty files, so the downstream
upload-coverage job keeps the real nodes' data intact.
* fix(ci): pre-create test-results in litellm_router_testing for empty-bucket reruns
litellm_router_testing also runs with parallelism 4. On a rerun of only the
failed tests, a node can receive no tests, so the test command never creates
test-results and the final store_test_results step can fail on the missing
path. Pre-create the directory up front, matching what local_testing_part1
and part2 already do and CircleCI's own guidance for parallel reruns.
* test(openai): retry wildcard chat completion on transient OpenAI 500
build_and_test reddened on test_openai_wildcard_chat_completion when the
real gpt-3.5-turbo-0125 call returned an OpenAI 500 ("The server had an
error while processing your request"). The base branch passed the same
call concurrently, so the 500 is an intermittent OpenAI server error, not
a regression. Add the same pytest-retry marker the sibling real-call tests
in this file already use so a transient upstream 500 no longer fails CI.
* fix duplicate cost callbacks for anthropic streaming pass-through
Two bugs caused _PROXY_track_cost_callback to see stream=True +
complete_streaming_response=None on every streaming pass-through request,
making the dedup guard in dispatch_success_handlers permanently inactive:
1. pass_through_endpoints.py created the Logging object with stream=False
for all requests. _is_assembled_stream_success short-circuits on
self.stream is not True, so has_dispatched_final_stream_success was
never set and any second dispatch went through unchecked.
Fix: set logging_obj.stream = True after stream detection.
2. _create_anthropic_response_logging_payload set complete_streaming_response
inside the try block after litellm.completion_cost(), so a pricing error
caused an early return without setting it on model_call_details.
Fix: set complete_streaming_response before the try block.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix stream
* add stream to logging obj
* test(pass_through): give mock logging object a real model_call_details dict
The anthropic passthrough logging payload now records the assembled
response on model_call_details before cost calculation, which requires
model_call_details to support item assignment. In production it is always
a dict; the existing unit test stubbed the logging object with a bare Mock
whose attribute is not subscriptable, so the new assignment raised
TypeError. Use a real dict to match the production logging object.
* test(pass_through): cover streaming logging-obj stream flag
The streaming branch of pass_through_request that marks the logging object
as streaming (logging_obj.stream and model_call_details["stream"]) had no
unit coverage, so the patch coverage gate flagged it. Add a regression test
that drives a streaming pass-through request through pass_through_request and
asserts the logging object is flagged as a stream before dispatch.
* test(pass_through): cover SSE-response stream flag fallback branch
The auto-detected streaming branch of pass_through_request (when a request
that was not flagged as streaming returns a text/event-stream response) sets
logging_obj.stream and model_call_details["stream"] but had no unit coverage,
so the codecov patch gate failed at 60%. Drive a non-streaming pass-through
request whose upstream response is SSE through pass_through_request and assert
the logging object is flagged as a stream before dispatch.
* fix(pass_through): gate complete_streaming_response on stream flag
perform_redaction only scrubs complete_streaming_response when
model_call_details["stream"] is True. Setting it unconditionally for
non-streaming Anthropic pass-through responses left the assembled
response unredacted in model_call_details, which is handed to logging
callbacks as kwargs when message logging is disabled. Only record it for
actual streaming responses so redaction always applies.
---------
Co-authored-by: mubashir1osmani <mubashir.osmani777@gmail.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(arize): enrich OpenInference attributes for better span rendering
Pure rendering enhancements to the Arize / Arize Phoenix integration. No
existing attribute keys or values are removed or overwritten; every new
emit is independently try/except-wrapped and fires only when its source
data is present so existing behavior is preserved.
What this adds
- Coerce non-dict response objects (e.g. httpx.Response from passthrough
routes) via JSON decode so id/model/usage extraction stops crashing
with "'Response' object has no attribute 'get'". Dicts and Pydantic
objects with .get pass through unchanged.
- Set OPENINFERENCE_SPAN_KIND defensively early so a downstream failure
can't blank the kind; the original late write (incl. TOOL upgrade) is
preserved.
- Add "passthrough" keyword to _infer_open_inference_span_kind so
allm_passthrough_route / llm_passthrough_route resolve to LLM instead
of UNKNOWN.
- Emit cache token breakdown: LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_READ /
_CACHE_WRITE / _AUDIO. Sources covered: OpenAI prompt_tokens_details
and Anthropic / Bedrock cache_{read,creation}_input_tokens.
- Render assistant tool_calls on both input and output messages via
MESSAGE_TOOL_CALLS.* (Pydantic-aware, handles ModelResponse choices).
Tool-result input messages also get MESSAGE_TOOL_CALL_ID and
MESSAGE_NAME.
- Render multimodal list-shaped content via MESSAGE_CONTENTS.* (OpenAI
image_url, Anthropic source.{media_type,data} as data: URI). Legacy
MESSAGE_CONTENT write is unchanged.
- Emit SESSION_ID (end_user_id / trace_id), USER_ID (only when not
already set by optional_params.user or model_params.user), and
litellm.{team_id,team_alias,key_alias} from StandardLoggingPayload
metadata.
- Emit llm.response.cost as float from StandardLoggingPayload.response_cost.
- Bedrock / Anthropic passthrough normalization: extract input from
additional_args.complete_input_dict and output from the coerced
provider response so INPUT_VALUE / OUTPUT_VALUE / LLM_INPUT_MESSAGES /
LLM_OUTPUT_MESSAGES are populated. Only runs when call_type contains
"passthrough" / "pass_through".
Tests
- 15 new unit tests covering each addition plus explicit regression
guards (USER_ID overwrite protection, passthrough normalizer scope,
coerce identity for dicts/.get-bearing objects, no spurious cache
emits).
- Existing test_arize_set_attributes count bumped from 26 to 27 to
account for the additional defensive span.kind write (same value,
written twice).
- tests/test_litellm/integrations/arize/: 70 passed (55 baseline + 15
new). tests/test_litellm/integrations/test_opentelemetry.py: 221
passed.
Co-authored-by: Cursor <cursoragent@cursor.com>
* refactor(arize): collapse additive try/except blocks into _safe_emit helper
The additive attribute emitters all share the same shape: run a callable,
swallow any exception to debug log so it cannot blank the span. Hoisting
that pattern into a single _safe_emit(label, fn, *args, **kwargs) helper
removes 5 repeated try/except blocks. Behavior unchanged; arize test
suite still passes (70/70).
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(arize): emit cost under canonical llm.cost.total key
Arize's "Total Cost" column reads the OpenInference-standard
`llm.cost.total` attribute. The previous custom `llm.response.cost`
key never surfaced in the trace list. Now emits both keys (canonical +
legacy) so renderers + any existing consumers both work.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(arize): keep span.kind=LLM for tool-using completions + render tool_calls in Output
A chat completion that passes `tools=[...]` or returns `tool_calls` is still
an LLM call per the OpenInference spec — TOOL is reserved for actual tool
execution. The previous override demoted these to TOOL, breaking Arize's
LLM-scoped dashboards/evals and skewing token/cost analytics for any
tool-using traffic.
Additionally, when an assistant response had no text content but did
request tool calls, `output.value` was set to the empty string so Arize's
"Output" pane rendered blank. Now serializes the tool_calls into a compact
JSON summary in `output.value` (the structured `MESSAGE_TOOL_CALLS.*`
attributes are still emitted unchanged).
Cleanups:
- extract `_get_tool_calls` and `_normalize_tool_call` helpers,
deduplicating the dict-vs-Pydantic + function-dict logic across
`_set_choice_outputs`, `_emit_message_tool_calls`, and the new
`_summarize_tool_calls_for_output`.
- drop redundant late `OPENINFERENCE_SPAN_KIND` write — the defensive
early write is now the single source of truth.
- remove a dead local re-import of `MessageAttributes`/`SpanAttributes`.
Tests: 73 pass (added regression guard asserting span.kind stays LLM for
completions that pass tools AND return tool_calls; existing call_count
assertion restored to 26).
Co-authored-by: Cursor <cursoragent@cursor.com>
* chore(arize): tighten cleanup — fold _get_tool_calls into _safe_get
Two tiny cleanups, no behavior change:
- collapse `_get_tool_calls` to use `_safe_get`, removing a 7-line
hand-rolled dict-vs-attribute fallback that duplicated existing logic.
- trim the `_set_choice_outputs` tool-call summary comment from 4 lines
to 2 (was over-explaining).
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(arize): address Greptile review — drop session_id=trace_id fallback, remove dead code, fix Black
Three Greptile-flagged issues + the Black formatting CI failure.
1. SESSION_ID no longer falls back to trace_id. Previously every span
without an explicit `user_api_key_end_user_id` would have its
session.id set to the per-request trace_id, which creates one
distinct "session" per request and breaks Arize's Session-grouping
analytics. Now SESSION_ID is emitted only when an explicit end-user
identifier exists, and the trace_id is emitted under its own
`litellm.trace_id` key so spans remain filterable by trace.
2. Removed dead `ArizeOTELAttributes.set_response_output_messages`
override. Confirmed zero callers in the entire repo (the live path
is `_set_choice_outputs` via `_set_response_attributes`). The
override was preexisting dead code, but the expansion of
`_set_choice_outputs` in this PR made the divergence misleading.
3. Removed permanently-dead first branch in cache_write detection.
`_safe_get(prompt_token_details, "cache_creation_tokens")` looks
for a key that neither OpenAI's `prompt_tokens_details` nor
Anthropic's payload ever exposes. Now reads straight off `usage`
for `cache_creation_input_tokens`.
4. Reformatted both files under Black 26.3.1 (the version CI uses
via `uv sync --frozen`). Local previously used 24.10.0.
Tests: 74/74 pass in the arize suite (added
`test_arize_does_not_use_trace_id_as_session_id_fallback`).
Combined arize + opentelemetry suite: 295/295 pass.
End-to-end verified live: tool-call still emits `span.kind=LLM` and
JSON tool_calls in `output.value`; `session.id` is now correctly
unset when no end_user_id is provided; `litellm.trace_id` is
populated; Bedrock passthrough input/output unchanged.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(arize): gate passthrough prompt export on message redaction
- Skip the complete_input_dict bridge in _maybe_normalize_passthrough when
should_redact_message_logging() is true, so enabling redaction no longer
leaks raw passthrough prompts into Arize (Veria security finding).
- Split passthrough input/output rendering into helpers to satisfy PLR0915.
- Remove dead call_type assignment (F841).
Validated live against a Bedrock passthrough proxy exporting to Arize:
non-redacted renders the real prompt on litellm_request; global
turn_off_message_logging yields input.value=redacted-by-litellm with the
raw_gen_ai_request child span suppressed and no SSN/marker leakage.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
* Add support for websocket via codex
* Add model alias and creds support
* fix: skip cost tracking for WS session wrapper call types
The @client decorator on _aresponses_websocket fires async_success_handler
with result=None after the session ends. This triggered cost tracking errors
because standard_logging_object is never built for None results.
Per-turn costs are correctly tracked by individual litellm.aresponses calls
inside the session. The outer session-level logging obj should not attempt
cost tracking.
Fix: skip _aresponses_websocket and _arealtime call types in deployment_callback_on_success,
RouterBudgetLimiting.async_log_success_event, and _PROXY_track_cost_callback.
* fix: address Greptile review comments
Fix JSON injection: use json.dumps instead of f-string interpolation for model name in WS body.
Add 30s timeout for first WS frame to prevent unbounded connection resource tie-up.
Restore per-event model override in streaming_iterator; fall back to connection-level model when event omits it.
Strengthen regression test: inject alias into kwargs via _update_kwargs_with_deployment mock so the test would fail on un-fixed code.
* fix: handle nested response.create format in first-frame model extraction
When ?model= is omitted, the first WS frame can carry the model in either flat
format (first_event["model"]) or nested format (first_event["response"]["model"]).
The flat-only check would silently reject clients using the nested wire format.
Mirrors the same two-format logic in _build_base_call_kwargs.
* fix: don't force connection-level custom_llm_provider on per-event model overrides
If a client sends a different model per response.create turn, litellm needs to
re-resolve the provider from that model string. Forcing the connection-level
custom_llm_provider would silently route the request to the wrong backend.
Only inject custom_llm_provider when the per-event model matches the
connection-level model.
* refactor: extract WS model extraction into testable function
Pull the flat/nested model extraction into _extract_model_from_first_ws_event
so tests import and exercise the real function rather than a copy.
* fix: compare providers not full model strings in _inject_credentials
The model == self.model guard was too strict: same-provider model variants
(e.g., vertex_ai/gemini-2.0 -> vertex_ai/gemini-1.5 on one connection) would
lose custom_llm_provider, breaking routing when a custom api_base is in use.
Compare the provider extracted by get_llm_provider instead, so same-provider
variants still inherit the connection-level provider while cross-provider
overrides let litellm re-resolve.
* style: black formatting
* refactor: extract first-frame model resolution to fix PLR0915 (too many statements)
* Fix responses WebSocket first-frame validation
* fix: classify WS first-frame read errors and clarify cost-skip log
Distinguish client disconnects from server errors when reading the
responses WebSocket first frame, make the cost-tracking skip log message
accurate for session wrappers (which do carry a model), and resolve the
connection-level provider once per session instead of on every
response.create event.
* test: cover WS first-frame read errors and same-provider credential injection
Adds regression tests for the still-uncovered responses WebSocket paths:
the timeout, invalid-JSON and missing-model branches of
_read_ws_model_from_first_frame, plus the provider comparison in
ManagedResponsesWebSocketHandler._same_provider and _inject_credentials
(same-provider model variants keep the connection provider; cross-provider
models re-resolve).
* fix(responses-ws): fall back to explicit custom_llm_provider when connection model is unresolvable
When a WebSocket session is opened with a custom deployment alias that litellm
cannot resolve to a provider, _connection_provider was None, so _same_provider
returned False for every resolvable per-event model and the connection-level
custom_llm_provider was dropped. Use the explicitly-set custom_llm_provider as
the connection provider in that case so same-provider per-event models still
inherit it while genuinely cross-provider models continue to re-resolve.
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(vertex): strip output_config.effort for models that reject it
Haiku 4.5 on Vertex AI does not support output_config.effort and 400s with
"output_config.effort: Extra inputs are not permitted". PR #27074 emptied
VERTEX_UNSUPPORTED_OUTPUT_CONFIG_KEYS so effort would forward for Opus/Sonnet
4.6+, but that made the strip unconditional across every Vertex Anthropic
model, including ones that don't support it. Claude Code injects effort into
its default Messages payload, so `claude --model claude-haiku-4.5` started
failing.
Make the sanitizer model-aware: drop output_config.effort for models that
don't advertise output_config support (or any reasoning effort level) while
forwarding it for those that do. The fix covers both the chat-completion and
Messages pass-through transformation paths since they share the helper.
* chore(vertex): log at debug when dropping unsupported output_config.effort
Operators pointing an unregistered Vertex Claude alias that does support
effort would otherwise see it stripped with no signal. Debug level keeps it
out of normal logs since Claude Code sends effort on every request.
* ci: reproduce default-Windows wheel install to guard MAX_PATH
The existing using_litellm_on_windows job installs the project with
`uv sync`, an editable source install that never copies package files
into a deep site-packages path, so it cannot see the 260-char MAX_PATH
overflow that breaks `pip install litellm` on default Windows. The
content-filter benchmark fixtures have hit that limit three times
(#21941, #22039, #29536), each caught only after release.
This adds a guard to the same job that builds the wheel and installs it
the way an end user would: into a venv whose site-packages prefix is
padded to a realistic worst-case Windows length (~100 chars), then
asserts the install completes and litellm imports. Any packaged path
long enough to bust MAX_PATH at that prefix is reported up front, so the
check is deterministic regardless of the runner's long-path setting,
while the real install also covers failure modes a length heuristic
cannot (half-unpacked packages, reserved names, case collisions).
This commit is the guard only; on the current tree it correctly fails
because nine fixtures still exceed the limit. The rename that brings
them back under it follows on this branch.
* fix(packaging): shorten content-filter benchmark fixtures under MAX_PATH
The 10 content-filter benchmark result fixtures used the legacy
block_{topic}_-_contentfilter_({yaml}).json naming, up to 176 chars
inside the wheel, which busts the Windows 260-char MAX_PATH limit once
extracted under a realistic site-packages prefix and aborts
`pip install litellm` on default Windows.
Rename them to the short {topic}_cf.json scheme that
_save_confusion_results already emits today (it splits the label on the
em-dash and writes f"{topic}_cf"), matching the insults_cf.json and
investment_cf.json files fixed earlier. Re-running the eval suite now
regenerates these same short names rather than recreating the long ones.
This drops the longest packaged path from 176 to 128, so the guard added
in the previous commit goes from red to green with a 32-char margin.
* test(windows): tidy MAX_PATH guard per review
Close the wheel zip via a context manager rather than leaning on
refcount collection, and select the wheel under dist/ by newest mtime so
a stale artifact from an earlier build cannot be tested instead of the
one just produced. Also pin down the venv-depth formula with a short
note: the +2 is the separator joining the venv root to "Lib" plus the
trailing separator before the entry, which lands the simulated
site-packages prefix at exactly 100 chars.
* fix(anthropic/adapter): open thinking block for reasoning_content-only streaming chunks
The /v1/messages streaming content-block classifier (_translate_streaming_openai_chunk_to_anthropic_content_block) only recognized thinking_blocks. OpenAI-compatible reasoning backends (vLLM/SGLang reasoning parsers: DeepSeek-R1, Qwen3, gpt-oss, ...) populate reasoning_content with thinking_blocks=None, so the classifier fell through to a text block. The delta translator already emits thinking_delta for reasoning_content, so those deltas landed inside a text block and Anthropic streaming clients (Claude Code, SDK .stream()) silently dropped the chain-of-thought.
Mirror the reasoning_content fallback already present in the non-stream translator and the streaming delta translator so the classifier opens a thinking block. Adds a focused regression test.
* fix(anthropic/adapter): reach reasoning_content branch when thinking_blocks attr is absent
Delta deletes the thinking_blocks attribute when unset, so the prior nested check was unreachable for reasoning-only chunks (vLLM/SGLang). Make it a sibling elif so the content block is classified as thinking.
* test(proxy): stop component-allowlist test leaking DATABASE_URL into xdist peers
The component-allowlist test pins throwaway DATABASE_URL/LITELLM_MASTER_KEY
values at import time via os.environ so importing proxy_server doesn't need a
live database. Those values persisted for the whole pytest-xdist worker, so a
sibling test sharing the worker (test_key_rotation_e2e's DB-backed E2E case)
saw the leaked sqlite DATABASE_URL, treated it as an available database instead
of skipping, and the Prisma engine rejected the non-postgres URL (P1012 ->
httpx.ConnectError). Restore the prior environment after the import so the
throwaway values never escape the module.
---------
Co-authored-by: Tai An <antai12232931@outlook.com>
* Fix error code and context id injection bug
* Add support for all A2A methods
* Add logging
* address greptile review: relay upstream JSON-RPC errors, move _PASCAL_TO_WIRE to module level, add error path tests
* fix(a2a): run pre_call_hook for tasks/resubscribe SSE path to enforce guardrails
tasks/resubscribe was returning the raw SSE stream without calling proxy_logging_obj.pre_call_hook, silently bypassing any guardrails configured on the agent. This patch calls pre_call_hook before streaming begins and wires post_call_failure_hook into the SSE generator so errors are logged. Adds a regression test verifying the hook is called.
* fix(a2a): use get_async_httpx_client instead of creating httpx clients per request
Creating httpx.AsyncClient instances per-request adds ~500ms latency. Switch _forward_jsonrpc and _forward_jsonrpc_sse to use the shared client from get_async_httpx_client(httpxSpecialProvider.A2A).
* fix(a2a): forward caller identity headers on task ops; validate push notification URL
Two security fixes for task management methods:
1. All task operations (tasks/get, tasks/list, tasks/cancel, tasks/resubscribe, push notification config methods) now forward X-LiteLLM-User-Id and X-LiteLLM-Team-Id headers to the upstream agent, so the agent can scope task access to the authenticated caller.
2. tasks/pushNotificationConfig/set validates the callback URL before forwarding: requires HTTPS and rejects private/loopback/reserved IP ranges and localhost hostnames to prevent SSRF.
* Fix A2A task hook and push URL handling
* fix(a2a): fix mypy type errors for request_id and header_name dict key types
* Fix A2A request id and params forwarding
* Forward trace IDs for A2A task calls
* fix(a2a): strip client-forwarded X-LiteLLM-* headers before applying authenticated identity
A client could send x-a2a-<agent>-x-litellm-user-id in their request and have it forwarded to the upstream agent as an authenticated identity header. Fix: sanitize any X-LiteLLM-* headers from agent_extra_headers before merging, then apply the authenticated identity headers last so they always override client-supplied values.
* Fix A2A SSE fallback JSON-RPC error code
* Fix A2A SSE error id backfill
* fix(a2a): validate both push notification url fields to close SSRF bypass
* fix(a2a): widen request_id annotation to match JSON-RPC id call sites
* fix(a2a): run post-call streaming hook for tasks/resubscribe so agent guardrails apply
tasks/resubscribe returned the raw upstream SSE stream without routing events
through the post-call streaming hook, so output guardrails configured on the
agent were silently skipped for streaming task subscriptions while every other
task method and message/stream applied them. Parse upstream JSON-RPC SSE events
and feed them through async_streaming_data_generator, matching message/stream,
so guardrails inspect the streamed task content. Adds a regression test that
fails when the streamed events bypass the guardrail hook.
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* Fix incorrect agent API request example payload structure (#29556)
* fix(otel): add litellm_metadata fallback in _get_span_context and _end_proxy_span_from_kwargs (#29427)
* fix(otel): add litellm_metadata fallback in _get_span_context and _end_proxy_span_from_kwargs
On /v1/messages and other LITELLM_METADATA_ROUTES, the parent OTel span
is stored in litellm_params['litellm_metadata'] instead of
litellm_params['metadata']. When the request body contains a native
'metadata' field (e.g. Anthropic's {"user_id": "..."}),
litellm_params['metadata'] gets overwritten and the parent span is lost,
producing orphan root spans with a different trace_id.
Add fallback checks to litellm_metadata in:
- _get_span_context(): so child spans find the correct parent
- _end_proxy_span_from_kwargs(): so the proxy span gets closed
Fixes: https://github.com/BerriAI/litellm/issues/27934
* test(otel): tighten assertions per Greptile review
- test_span_context_metadata_takes_priority: assert litellm_metadata
span is never accessed, proving metadata takes priority
- test_span_context_no_parent_when_neither_has_span: assert both ctx
and detected_span are None
---------
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Co-authored-by: yuneng-jiang <yuneng@berri.ai>
Co-authored-by: Aneesh-Fiddler <aneeshfiddler@gmail.com>
Co-authored-by: Sameer Kankute <sameer@berri.ai>
* fix: remove premature end-user budget check from get_end_user_object (#29420)
* fix(proxy): remove premature end-user budget check from get_end_user_object
Problem:
- `_check_end_user_budget()` was called inside `get_end_user_object()`
- This caused budget checks to run BEFORE `skip_budget_checks` could be evaluated
- Zero-cost models (e.g., local vLLM) were incorrectly blocked when
end-users exceeded their budget, even though they should bypass budget checks
Solution:
- Remove `_check_end_user_budget()` calls from `get_end_user_object()`
- Budget enforcement now happens exclusively in `common_checks()` where
`skip_budget_checks` context is available
- `get_end_user_object()` keeps `route` as optional in function parameter for backwards compatibility and future implementation.
* refactor(tests): update budget enforcement tests to reflect changes in get_end_user_object
- test_get_end_user_object() verifies data fetching
- test_check_end_user_budget() verifies enforcement
- test_budget_enforcement_blocks_over_budget_users() integrates _check_end_user_budget()
- test_resolve_end_user_reraises_budget_exceeded() is now test_resolve_end_user since no budget exceeded is thrown in get_end_user_object()
* Gemini /images/generate and /images/edits billing fixes + add support for size and aspect ratio params (#29534)
* Fix Gemini image config mapping
* Address Gemini image config review
* Format Gemini image generation transform
* Fix Gemini image token usage logging
* Share Gemini image request helpers
* Fix Gemini Imagen model routing
* Fixes as per self code review
* Fixes per internal code review
* Stop gating Imagen imageSize forwarding
* Document Gemini image size mapping source
* chore: retrigger lint
* Clarify Gemini candidate count precedence
* Add Inception provider (#29522)
* add inception as provider (chat, fim)
* linting
* seperate test suite for chat and fim
* fix test coverage
* fix: model hub custom pricing model info (#29293)
* Opik user auth key metadata extractors (#28397)
* fix: enhance Opik metadata extraction to include user API key auth context fixed after refactoring to extractor logic
* test: add unit tests for OPik metadata extraction logic
* fix: enhance extract_opik_metadata function to prioritize metadata sources for improved accuracy
* fix(ci): clarified comments and edited unit tests
* test: add unit tests for OPik metadata extraction with auth and requester overrides
* fix(ui): replace fixed favicon.ico with current api get /get_favicon (#29532)
Signed-off-by: José Luis Di Biase <josx@interorganic.com.ar>
* fix(vertex/gemini): keep tool_call reference when a text-only assistant message follows (#29561)
`_gemini_convert_messages_with_history` tracks `last_message_with_tool_calls`
so a following tool result can be matched back to its tool call. The assignment
was inside a branch guarded by
`assistant_msg.get("tool_calls", []) is not None`, which is also True for a
text-only assistant message (an empty list is not None). As a result, an
assistant message with no tool calls that appears between a tool call and its
tool result overwrote the reference, and conversion failed with:
Exception: Missing corresponding tool call for tool response message.
This shape is common: a model emits a short narration/assistant message after a
tool call before the tool result is appended.
Only update `last_message_with_tool_calls` when the assistant message actually
carries tool_calls (or a function_call). Adds a regression test.
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Co-authored-by: yuneng-jiang <yuneng@berri.ai>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* Add 1-hour cache write pricing for EU/AU/JP Bedrock Anthropic models (#28572)
* fix(thinking): handle None thinking param in is_thinking_enabled (#28598)
Squash-merged by litellm-agent from Terrajlz's PR.
* feat(helm): support tpl rendering in podAnnotations (#28609)
Squash-merged by litellm-agent from devauxbr's PR.
* Forward custom_llm_provider through the Responses API bridge (Fixes#28505) (#28575)
* Forward custom_llm_provider through the Responses API bridge (Fixes#28505)
When a Chat Completions request to a GPT-5.4+ model contains both
`tools` and `reasoning_effort`, `completion()` auto-routes through
`responses_api_bridge`. The bridge handler called
`litellm.responses()` / `litellm.aresponses()` without forwarding the
already-resolved `custom_llm_provider`, so the downstream call
re-invoked `get_llm_provider()` with `custom_llm_provider=None` and
stripped a second provider prefix from a `provider/provider/model`
deployment string.
For a deployment configured as `openai/openai/openai/gpt-5.5`,
the bridge flow sent `openai/gpt-5.5` to the upstream API instead of
the correct `openai/openai/gpt-5.5`. Upstream APIs that enforce
model-name allow-lists rejected this as `key_model_access_denied`.
Fix: pass the locally-resolved `custom_llm_provider` into both the
sync `responses()` and async `aresponses()` calls so the downstream
`_resolve_model_provider_for_responses` sees an explicit provider
and skips the second prefix-strip.
New regression test
`tests/test_litellm/completion_extras/test_responses_bridge_provider_propagation.py`
pins both call sites: each must forward `custom_llm_provider`.
* fix(28505): set custom_llm_provider on request_data instead of as duplicate kwarg
Greptile flagged that the previous patch passed custom_llm_provider as an
explicit kwarg to responses()/aresponses() while request_data already
carried it via the spread of sanitized_litellm_params, which would raise
TypeError: got multiple values for keyword argument on every real bridge
call.
Switches to assigning request_data['custom_llm_provider'] before the call
so the resolved provider wins over whatever sanitized_litellm_params spread
in, without duplicating the kwarg.
Updates the regression test to seed request_data with a sentinel
custom_llm_provider so it actually exercises the overwrite path (the
previous test mocked transform_request with a minimal dict and never hit
the conflict).
* chore: trigger shin-agent re-eval on retargeted staging base
* chore: trigger shin-agent re-eval against updated Greptile state
* Add 1-hour cache write pricing for EU/AU/JP Bedrock Anthropic models
The 1-hour prompt-cache write tier
(`cache_creation_input_token_cost_above_1hr`) was added to the
us./global. variants of the Claude 4.5/4.6/4.7 family on Bedrock, but
the eu./au./jp. cross-region inference profiles were left without it.
AWS Bedrock pricing applies the same +10% regional premium across all
geo profiles, so eu./au./jp. should carry the same 1-hour rates as
us. (1.6x the 5-minute regional rate).
Without these fields, cost tracking on EU/AU/JP Bedrock 1-hour-TTL
prompt caching falls back to the 5-minute write rate and undercounts
spend by ~60% for European, Australian, and Japanese tenants.
Adds the 1-hour tier (and Sonnet 4.5's long-context >200K tier where
AWS publishes one) to 14 regional Bedrock entries in both
`model_prices_and_context_window.json` and the bundled
`model_prices_and_context_window_backup.json`:
- eu./au. Opus 4.6 ($11.00 / MTok)
- eu./au. Opus 4.7 ($11.00 / MTok)
- eu./au./jp. Sonnet 4.6 ($6.60 / MTok)
- eu./au./jp. Sonnet 4.5 ($6.60 / MTok regular, $13.20 / MTok LC)
- eu./au./jp. Haiku 4.5 ($2.20 / MTok)
Also extends `tests/test_litellm/test_bedrock_anthropic_1hr_cache_pricing.py`
with a `REGIONAL_EXPECTED` parametrized block covering all 13 new
entries plus the existing 1.6x ratio invariant.
Note: `eu.anthropic.claude-opus-4-5-20251101-v1:0` carries the
wrong 5m rate today (base 6.25e-06 instead of regional 6.875e-06),
which would break the 1.6x ratio check. It is intentionally left out
of this PR so the scope stays "1-hour cache tier addition" — a
separate follow-up should correct the EU 5m rates for Opus 4.5.
---------
Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: Sameer Kankute <sameer@berri.ai>
* Add 1-hour cache write pricing tier for Vertex AI Anthropic models (#28569)
* fix(thinking): handle None thinking param in is_thinking_enabled (#28598)
Squash-merged by litellm-agent from Terrajlz's PR.
* feat(helm): support tpl rendering in podAnnotations (#28609)
Squash-merged by litellm-agent from devauxbr's PR.
* Forward custom_llm_provider through the Responses API bridge (Fixes#28505) (#28575)
* Forward custom_llm_provider through the Responses API bridge (Fixes#28505)
When a Chat Completions request to a GPT-5.4+ model contains both
`tools` and `reasoning_effort`, `completion()` auto-routes through
`responses_api_bridge`. The bridge handler called
`litellm.responses()` / `litellm.aresponses()` without forwarding the
already-resolved `custom_llm_provider`, so the downstream call
re-invoked `get_llm_provider()` with `custom_llm_provider=None` and
stripped a second provider prefix from a `provider/provider/model`
deployment string.
For a deployment configured as `openai/openai/openai/gpt-5.5`,
the bridge flow sent `openai/gpt-5.5` to the upstream API instead of
the correct `openai/openai/gpt-5.5`. Upstream APIs that enforce
model-name allow-lists rejected this as `key_model_access_denied`.
Fix: pass the locally-resolved `custom_llm_provider` into both the
sync `responses()` and async `aresponses()` calls so the downstream
`_resolve_model_provider_for_responses` sees an explicit provider
and skips the second prefix-strip.
New regression test
`tests/test_litellm/completion_extras/test_responses_bridge_provider_propagation.py`
pins both call sites: each must forward `custom_llm_provider`.
* fix(28505): set custom_llm_provider on request_data instead of as duplicate kwarg
Greptile flagged that the previous patch passed custom_llm_provider as an
explicit kwarg to responses()/aresponses() while request_data already
carried it via the spread of sanitized_litellm_params, which would raise
TypeError: got multiple values for keyword argument on every real bridge
call.
Switches to assigning request_data['custom_llm_provider'] before the call
so the resolved provider wins over whatever sanitized_litellm_params spread
in, without duplicating the kwarg.
Updates the regression test to seed request_data with a sentinel
custom_llm_provider so it actually exercises the overwrite path (the
previous test mocked transform_request with a minimal dict and never hit
the conflict).
* chore: trigger shin-agent re-eval on retargeted staging base
* chore: trigger shin-agent re-eval against updated Greptile state
* Add 1-hour cache write pricing tier for Vertex AI Anthropic models
GCP Vertex AI publishes a separate 1-hour cache write column for the
Claude family (1.6x the 5-minute write rate, matching the documented
Bedrock ratio). LiteLLM's Vertex AI Anthropic entries only carry the
5-minute tier, so any request that uses `cache_control: {"ttl": "1h"}`
on Vertex AI Claude is undercounted in cost tracking by ~60%.
The runtime side already supports the 1-hour tier — `VertexAIAnthropicConfig`
extends `AnthropicConfig`, populating `ephemeral_1h_input_tokens`, and
`_calculate_cache_creation_cost` reads `cache_creation_input_token_cost_above_1hr`.
Only the price registry was missing data.
Adds the field to 19 vertex_ai/claude-* entries across both
`model_prices_and_context_window.json` and the bundled
`model_prices_and_context_window_backup.json`:
- Haiku 4.5 ($1.25 -> $2.00 / MTok)
- Sonnet 3.7 / 4 / 4.5 / 4.6 ($3.75 -> $6.00 / MTok)
- Opus 4.5 / 4.6 / 4.7 ($6.25 -> $10.00 / MTok)
- Opus 4 / 4.1 ($18.75 -> $30.00 / MTok)
Adds `tests/test_litellm/test_vertex_anthropic_1hr_cache_pricing.py`
mirroring the Bedrock equivalent — pins each (5m, 1h) pair per model
and asserts the 1.6x ratio across the family.
Fixes#27781.
---------
Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: Sameer Kankute <sameer@berri.ai>
* Fix Gemini multimodal function responses (#29325)
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Co-authored-by: yuneng-jiang <yuneng@berri.ai>
* address greptile review: add _transform_image_usage method and model-map supports_image_size flag
- Add _transform_image_usage instance method to GoogleImageGenConfig that
delegates to transform_gemini_image_usage, fixing the regression test
- Replace hardcoded "2.5-flash" string check in supports_gemini_image_size
with a get_model_info lookup on supports_image_size (default true)
- Add supports_image_size: false to all gemini-2.5-flash model entries in
model_prices_and_context_window.json so capability is controlled via the
model map rather than embedded in code
* fix test failures: schema validation, mypy type, model info plumbing, pricing test
- Add supports_image_size to ModelInfoBase TypedDict so get_model_info surfaces it
- Pass supports_image_size through _get_model_info_helper constructor call
- Fix supports_gemini_image_size to use value is not False (None means unset, defaults to True)
- Add supports_image_size to JSON schema in test_aaamodel_prices_and_context_window_json_is_valid
- Correct gemini-3.1-flash-lite pricing assertions in test to match JSON values
* Add Azure AI Kimi K2.6 metadata (#27052)
* Add Azure AI Kimi K2.6 metadata
* Scope Kimi metadata test cost map setup
* fall back to substring check for models not in model_prices_and_context_window.json
Models like gemini-2.5-flash-image-preview are not in the pricing JSON,
so get_model_info raises. Fall back to "2.5-flash" not in model when the
JSON has no explicit supports_image_size entry for the model.
* fix(inception): don't forward global litellm.api_key to Inception FIM
Match the Inception chat config: resolve only an Inception-specific key
(param, litellm.inception_key, or INCEPTION_API_KEY) for the text-completion
FIM path. The global litellm.api_key (often an OpenAI key) was both leaking
to api.inceptionlabs.ai and taking precedence over the configured Inception
key when set.
* fix(auth): enforce end-user budget on custom-auth path that skips common_checks
get_end_user_object() no longer raises BudgetExceededError, so custom-auth
deployments with custom_auth_run_common_checks unset (which skip the
centralized common_checks gate) stopped enforcing the end-user budget,
letting an over-budget end user keep making requests. Re-enforce the
budget in _run_post_custom_auth_checks on that path.
---------
Signed-off-by: José Luis Di Biase <josx@interorganic.com.ar>
Co-authored-by: Isha <72744901+IshaMeera@users.noreply.github.com>
Co-authored-by: aneeshsangvikar <aneeshsangvikar@fiddler.ai>
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Co-authored-by: yuneng-jiang <yuneng@berri.ai>
Co-authored-by: Aneesh-Fiddler <aneeshfiddler@gmail.com>
Co-authored-by: Suleiman Elkhoury <108065141+suleimanelkhoury@users.noreply.github.com>
Co-authored-by: Dmitriy Alergant <93501479+DmitriyAlergant@users.noreply.github.com>
Co-authored-by: Yanis Miraoui <yanis.miraoui19@imperial.ac.uk>
Co-authored-by: Lovro Seder <vrovro@gmail.com>
Co-authored-by: Thomas Mildner <12685945+Thomas-Mildner@users.noreply.github.com>
Co-authored-by: José Luis Di Biase <josx@interorganic.com.ar>
Co-authored-by: Lai Quang Huy <64073540+1qh@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Filippo Menghi <113345637+Cyberfilo@users.noreply.github.com>
Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: ZHONG Ziwen <67355585+zzw-math@users.noreply.github.com>
Co-authored-by: Emerson Gomes <emerson.gomes@thalesgroup.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* test(pass-through): move Gemini pass-through tests to gemini-3.1-flash-lite
gemini-2.5-flash-lite is a generation behind and is slated for discontinuation on Vertex AI no earlier than October 16, 2026, so the pass-through suite was exercising an aging model. Every reference now points at gemini-3.1-flash-lite, which is GA and already priced in the cost map so the spend-logging assertions still compute a real cost
test_vertex.test.js also gains jest.retryTimes(3) to match the sibling spend tests. The CI failures were intermittent 429 RESOURCE_EXHAUSTED from Vertex quota pressure, and that file was the only one without a retry, so a single rate-limited request was failing the whole job
* test(pass-through): point Vertex tests at the global endpoint for gemini-3.1-flash-lite
gemini-3.1-flash-lite is not served on the Vertex us-central1 regional endpoint for the CI project, so the Vertex pass-through tests were returning a deterministic 404 "Publisher Model ... was not found or your project does not have access to it" while the Gemini API tests passed. Move the Vertex clients to the global location, which the pass-through router maps to aiplatform.googleapis.com, where the 3.1 family is served
* fix(key_generate): allow team members to create keys on org-scoped teams
When a virtual key is created for a team, enterprise logic inherits the
team's organization_id onto the key (add_team_organization_id). Since the
VERIA-55 org-IDOR fix, /key/generate then required the caller to be an
explicit LiteLLM_OrganizationMembership member of that org, returning
403 "Caller is not a member of organization_id=<uuid>". Admins normally
only add users to teams (not orgs), so self-serve key creation regressed
for any user on an org-scoped team (regression since v1.84.0-rc.1).
Skip the org-membership check when organization_id was inherited from the
key's team (organization_id == team_table.organization_id). Team-level
authorization already gates this path, so team membership is sufficient.
The membership check still runs when a caller assigns an organization_id
that did not come from the key's team, preserving the IDOR protection.
Adds regression tests covering both the team-inherited (allowed) and
foreign-org (still blocked) cases.
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(key_generate): cover mismatched team org IDOR path on generate
Add test_generate_key_foreign_org_with_mismatched_team_still_enforces_membership
for the case where a team is present but request organization_id differs from
team_table.organization_id. Enterprise inheritance is no-op'd in the test so
the guard is exercised directly; membership validation must still run.
Addresses Greptile review on #29310.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(auth): let internal users view search tools
Internal users could not see search tools in the UI even when an admin
created them, while vector stores were visible. The Search Tools page
rendered but its list calls 403'd because the read routes were not in
internal_user_routes. Grant internal users read-only access to the
listing and provider routes; create/update/delete stay admin-only.
Resolves LIT-3150
* fix(auth): scope /search_tools/list to caller-allowed tools
Adding the read routes to internal_user_routes let any internal user
call /search_tools/list, which returned every configured tool's id,
provider, api_base, and metadata regardless of the caller's
object_permission.search_tools allowlist. The api_key was masked, so
this was metadata disclosure rather than a credential leak, but it
ignored the key/team scoping that /search already enforces.
Filter the listing through the same can_key_call_search_tool /
can_team_call_search_tool checks (exposed as a boolean
can_user_view_search_tool), so non-admin callers only see tools they
may invoke; admins still see all. Mirrors how /vector_store/list
scopes results.
* fix(types): resolve mypy errors in list_search_tools
The config and DB build loops reused one loop variable, so mypy pinned
it to the config element type (SearchToolTypedDict); its .get() calls
returned object and the DB element (SearchTool) failed the reuse
assignment. Give each loop its own variable so each gets its real type,
and coerce the config tool's SearchToolInfoTypedDict to a plain dict to
match SearchToolInfoResponse.search_tool_info.
* test(proxy/utils): pin ProxyLogging behavior
Add behavior-pinning tests for the ProxyLogging cluster in
litellm/proxy/utils.py under tests/test_litellm/proxy/utils/proxy_logging/.
Covers InternalUsageCache, _CallbackCapabilities, top-of-file helpers
(print_verbose, _get_email_logger_class, _accepts_litellm_call_info,
_enrich_http_exception_with_guardrail_context), the full ProxyLogging
class (lifecycle, MCP-LLM bridging, capability probes, guardrail
pipeline, pre/during/post/streaming hooks, alerting), plus the
bottom-of-region helpers (on_backoff, jsonify_object, _lookup_deprecated_key).
Each pinned symbol has happy-path and error-path coverage; happy paths
use direct dict-equality with three or more keys (or HiddenParams /
Pydantic model_validate where the surface is a Pydantic shape). The
subdirectory carries a local _pin_check.py and _coverage_check.py that
enforce the gate without surfacing numeric thresholds in CI logs.
Wires tests/test_litellm/proxy/utils into the existing test-path block
in .github/workflows/test-unit-proxy-endpoints.yml.
* test(proxy/utils): drop unused mock_httpx_client fixture
Declared in conftest.py but never referenced by any test. Removing
the dead fixture per Greptile P2 feedback.
* test(proxy/utils): drop local-only gate scripts from PR
_pin_check.py and _coverage_check.py are local stopping signals (not
wired into CI, consume a gitignored .pin_list.txt). They served their
purpose telling the engineer when to stop writing tests; the pytest
suite is the artifact that belongs in the repo.
---------
Co-authored-by: Claude <noreply@anthropic.com>
* test(proxy/utils): pin PrismaClient and spend-update behavior
PR2 of the litellm/proxy/utils.py behavior-pinning plan
(https://www.notion.so/37343b8acdab81f68f39f66915f62bcf). Adds
tests/test_litellm/proxy/utils/prisma_and_spend/, with happy + error
pins for every symbol in the PR2 list: the config-param cache,
PrismaClient lifecycle/data ops/engine watcher/reconnect/health
clusters, the user-row cache and SMTP helper, password/token helpers,
ProxyUpdateSpend, and the module-level spend functions.
Tests run against fully-mocked Prisma stacks (patched ``Prisma`` /
``PrismaWrapper`` at fixture setup), with a fake SMTP transport and a
clock-driven asyncio.sleep for the monitor loop, so unit runs need no
DB or network. ``_pin_check.py`` enforces happy + error coverage for
every symbol; ``_coverage_check.py`` filters branch + line coverage to
the PR2 source range (lines 2,668-5,541) and prints PASS / FAIL with
no numbers. Workflow shard ``tests/test_litellm/proxy/utils`` is added
to the existing proxy-endpoints job.
* test(proxy/utils): commit pin list and drop dead exclusion line
Addresses Greptile review feedback on PR #29488:
- Check in ``.pin_list.txt`` (force-added, overriding the repo-wide
``.gitignore`` rule) so reviewers can reproduce the ``_pin_check.py``
PASS shown in the PR description without first regenerating the
file from Notion.
- Remove the unreachable ``_harness_smoke_test.py`` continue in
``_pin_check.py``: the surrounding ``test_*.py`` glob already
excludes underscore-prefixed files; rephrase the docstring instead.
* test(proxy/utils): shift PR2 coverage line range by +1 after merge
``litellm_internal_staging`` added one line in ``ProxyLogging`` at
``utils.py:645`` (PR1 territory, before the PR2 region). Bump the
``_PR2_LINE_START`` / ``_PR2_LINE_END`` constants accordingly so the
coverage gate keeps scoring the same source region after the merge.
* test(proxy/utils): drop committed pin-list and gate scripts
``_pin_check.py``, ``_coverage_check.py``, and ``.pin_list.txt`` are
local-only stopping signals: no workflow or pytest collection invokes
them, so committing them adds rot risk (line-range drift in the
coverage check, pin-list staleness) without any enforcement upside.
The pin-list contract lives in the Notion plan; the tests themselves
are the durable artifact.
---------
Co-authored-by: Claude <noreply@anthropic.com>
* test(proxy/utils): pin bottom-of-file helper behavior
Pin current behavior of the bottom-of-file pure-function helpers in
litellm/proxy/utils.py (projection, team config, time helpers,
guardrail merge, error helpers, URL/path helpers, premium gate, model
access, and misc DB/API-key helpers).
Adds tests/test_litellm/proxy/utils/helpers/ with one happy + one error
test per pinned symbol; folds the prior single-test
tests/test_litellm/proxy/test_utils.py into test_url_helpers.py and
deletes the old file. _pin_check.py and _coverage_check.py serve as
local stopping gates.
Adds tests/test_litellm/proxy/utils to the existing test-path block in
.github/workflows/test-unit-proxy-endpoints.yml.
Plan: https://www.notion.so/37343b8acdab81f68f39f66915f62bcf
Pin list: https://www.notion.so/37343b8acdab8150acdbf40e5756869f
* test(proxy/utils): apply greptile fixes to behavior-pinning gates
Address findings from the sibling PR1/PR2 greptile reviews that also
apply to this PR:
- Commit pin_list.txt alongside the gate script (was previously a
gitignored .pin_list.txt fetched from Notion). The gate is now
reproducible without out-of-band setup.
- Resolve the coverage region by locating the first pinned symbol's
def line in litellm/proxy/utils.py at runtime, instead of hardcoded
line numbers that drift when lines above shift.
- Word-boundary the pin reference check so pins like update_spend do
not falsely match update_spend_logs_job.
- Drop the dead _harness_smoke_test.py exclusion; the test_*.py glob
already filters underscore-prefixed files.
* test(proxy/utils): drop local-only stopping-signal scripts
Remove _pin_check.py, _coverage_check.py, and pin_list.txt. These were
dev-time tooling for knowing when test authoring was done; they are
not wired into CI and the test files themselves are the merge artifact.
---------
Co-authored-by: Claude <noreply@anthropic.com>
Auth failures on management endpoints such as team/list and
organization/list (invalid or expired keys) were raised as
ProxyException, whose __str__ returned an empty string, so the OTEL
SERVER span recorded an error with no message. ProxyException now
stringifies to its message, get_error_information prefers the explicit
.message attribute, and the proxy exception handlers stamp a consistent
error.type, error.code and error.message on the span
Resolves LIT-3515
* fix(auth): preserve 401 status for expired JWTs in OTel traces
Expired JWT access tokens raised a generic Exception with no status
code attached. Because the codeless exception was logged to OTel via
post_call_failure_hook before auth_exception_handler re-wrapped it as
ProxyException(401), the OTel span never set http.response.status_code
and trace viewers displayed it as a generic 500. Clients still got a
401 back, so traces and actual responses diverged.
Raise ProxyException(code=401, type=expired_key) directly at the source
in both JWT decode paths so the 401 is consistent across the client
response and the OTel http.response.status_code attribute, matching how
virtual-key expirations are handled.
* fix(auth): preserve 401 for expired JWTs on issuer-scoped path
The issuer-scoped JWT path (_auth_jwt_with_issuer) still raised a generic
Exception on expiry, surfacing as a 500 in client responses and OTel traces.
Raise ProxyException with expired_key/401 there too, matching auth_jwt, and
add a regression test exercising the issuer path end-to-end
* fix(vertex_ai): use user-supplied api_base as is for Model Garden OpenAI-compat path
* chore(tests): url assertions and outputs
* fix(tests): fixing reference to unused test
* fix(aiohttp): drop octet-stream content-type on bodyless requests
The aiohttp transport forwarded httpx's empty request body straight
to aiohttp, which attaches a default Content-Type: application/octet-stream
for any bytes payload. Bodyless requests such as DELETE /responses/{id} then
hit OpenAI with that header and were rejected with unsupported_content_type,
breaking the e2e_openai_endpoints test_basic_response check. Coercing an
empty body to None makes aiohttp behave like the httpx transport and send no
content-type for bodyless requests.
---------
Co-authored-by: Steven Kessler <9701252+stvnksslr@users.noreply.github.com>
* test: stabilize batch VCR coverage
* test: replay bedrock batch s3 uploads
* test: stop batch tests leaking live uploads
* test: keep bedrock batch workflow off live s3
* test: mock bedrock batch workflow network
* test: accept realtime guardrail refusal wording
* test: update gemini thought signature model
* test: quiet logging worker atexit flush
* test: address Greptile review on batch VCR fixes
Handle content= bodies in the bedrock batch post stub so payload
extraction does not raise a TypeError when a request omits json and
data. Restore litellm list state faithfully by preserving None instead
of coercing it to an empty list, so callbacks that start as None are not
turned into [] after a test. Set logging.raiseExceptions inside the try
block in the atexit flush so the finally always restores the previous
value.
* test: scope atexit logging suppression to the drain loop
Wrap only the queue drain loop in LoggingWorker._flush_on_exit with the
logging.raiseExceptions toggle so the process-wide global is suppressed for
the smallest possible window, keeping other threads' logging error reporting
intact outside the loop.
* test: cover atexit flush error-swallow branch in LoggingWorker
The _flush_on_exit drain loop was wrapped in a try/finally to scope the
logging.raiseExceptions toggle, which reindented the existing edge-case
branches into the diff and dropped patch coverage below target. Add a
regression test that enqueues a coroutine which raises during the atexit
flush and asserts the failure is swallowed while later queued events are
still drained, exercising the silent-failure path directly.
* fix(tests): drop import-time completion call in test_register_model
test_update_model_cost_via_completion() was invoked at module scope, so it
ran during pytest collection and fired a live OpenAI completion. The local
test jobs glob the whole tests/local_testing folder and let pytest import
every file, narrowing what runs only afterward with -k, so this call
executed in every one of those jobs regardless of their filter. When the
request failed (for instance a 429 once the OpenAI account hit its quota),
collection of the file errored and aborted the entire session, which is why
langfuse, assistants, router and local_testing_part2 all reported
"ERROR collecting tests/local_testing/test_register_model.py" and never ran
their own tests.
Remove the stray call and add a regression that parses the module and fails
if any locally defined function is invoked at module scope again
* test: also guard async def from module-scope invocation
ast.AsyncFunctionDef is a distinct node from ast.FunctionDef, so an async
test invoked at module scope would have slipped past the guard. Collect
both kinds of definitions
* fix(responses): send Content-Type application/json on OpenAI responses requests
OpenAI's responses API now rejects body-less requests (GET/DELETE) that
arrive without a content type, returning 500 "Unsupported content type:
'application/octet-stream'. This API method only accepts 'application/json'
requests". litellm's create path got the header for free because httpx sets
it when a json body is present, but the delete/get handlers send no body and
so sent no content type. The official OpenAI SDK declares
Content-Type: application/json on every request; mirror that in
validate_environment so all OpenAI responses calls carry it.
This is what made tests/openai_endpoints_tests/test_e2e_openai_responses_api.py::test_basic_response
fail on the responses.delete() call.
* feat: route future Claude models to Anthropic provider via pattern matching
Add pattern-based matching for Claude model names so that future models
(e.g., claude-opus-4-9, claude-sonnet-5-0) are automatically routed to
the Anthropic provider without requiring model_prices_and_context_window.json
updates.
The pattern matches: claude-{opus|sonnet|haiku}-{major}-{minor}[-YYYYMMDD]
https://claude.ai/code/session_017asCVDN5jBFMBcZRjiQR6C
* fix: don't hard-code the tier names
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
* style: move import re to module level (PEP 8)
Move `import re` from inside the module body to the top-level imports
section, following PEP 8 style guidelines that all imports should
appear at the top of the file.
https://claude.ai/code/session_01Dt8fzn81eYMfxu1MoBa5hN
* test: fix claude-mini-4-5 assertion to match generic-tier pattern
The pattern intentionally accepts any [a-z]+ tier (see f835f84) rather
than a hard-coded opus|sonnet|haiku list, so claude-mini-4-5 routes to
anthropic and the old 'is False' assertion was wrong. Replace it with a
positive test that locks in the generic-tier behavior and guards against
a regression to hard-coded tier names.
* style: collapse _CLAUDE_PATTERN to one line (black)
Black 26.3.1 (CI) collapses the re.compile call onto a single line since
it fits within the line limit. Fixes the failing lint check.
---------
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
* fix(ui/agents): make A2A skill tags enterable and validated
Skill tags were marked required but rendered as a comma-split text input
that couldn't surface validation and let empty values save. Switch tags
and examples to Select tag inputs, drop the misleading "Required" skills
label (the API allows zero skills), and validate the full configure step
so an added skill must be complete before advancing.
Resolves LIT-3153
* fix(ui/agents): allow Enter to create skill tags/examples
Drop open={false} from the tags and examples Select inputs. With the
dropdown forced closed, AntD suppresses the "create from input" option,
so pressing Enter (as the placeholder instructs) did nothing. Matches the
existing extra_headers Select.
* feat(agents): add LangFlow agent provider with A2A session bridging
Register LangFlow as a completion provider and agent type (UI + /api/v1/run),
and map A2A contextId to LangFlow session_id for multi-turn conversations.
Co-authored-by: Cursor <cursoragent@cursor.com>
* docs(providers): document langflow in provider_endpoints_support.json
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(agents): address Greptile review for LangFlow integration
Move A2A contextId→session_id mapping into LangFlow A2A provider config,
add langflow.svg logo, remove live integration test, use model for token count.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(langflow): prevent flow_id override via request optional_params
Derive flow_id only from the authorized model name and reject flow_id
kwargs so callers cannot invoke a different LangFlow run endpoint.
Co-authored-by: Cursor <cursoragent@cursor.com>
* refactor(langflow): remove redundant flow_id branch in _get_flow_id
* fix(langflow): surface an error when the run response has no extractable message
Previously the response parser returned the raw JSON blob as the assistant
message when it could not find message text, silently presenting an
unparseable payload as a valid answer. It now returns None and the caller
raises a LangFlowError so the failure is visible to the client.
* fix(langflow): URL-encode flow_id path segment to prevent path injection
flow_id is taken from the model suffix and interpolated into
/api/v1/run/{flow_id}. Without path-segment encoding a model such as
langflow/../../x (or one containing ?) could move the request off the run
endpoint to another path on the configured LangFlow server using the
operator x-api-key. Encode the segment with quote(safe="") so it always
stays a single path segment.
* fix(langflow): reject empty flow_id from model name
* fix(langflow): return stripped flow_id so validation matches URL path
* fix(langflow): reject caller-supplied tweaks to prevent flow component override
* fix(langflow): reject caller-supplied tweaks injected via extra_body
The transform_request guard only inspected optional_params, but extra_body
is popped before transform_request runs and merged into the request body
afterward, letting a caller reintroduce tweaks and override the
operator-configured LangFlow flow components. Validate the final request
body in sign_request so tweaks cannot reach LangFlow through extra_body.
* test(langflow): move provider tests into mirrored coverage path
The langflow tests lived under tests/llm_translation/, whose CircleCI job
runs without --cov and uploads nothing to Codecov, so none of the new
langflow code counted toward patch coverage (codecov/patch reported 9.78%
of the diff hit against a 70.83% target).
Relocate them to tests/test_litellm/llms/langflow/, which the GitHub
Actions provider job runs with --cov=./litellm and uploads, and add
regression tests for the previously untested happy paths (transform_response
building the ModelResponse with usage, non-JSON body handling, last-user
message extraction, outputs-dict response shape, sign_request pass-through,
error class and stream flags). Patch coverage on the diff is now ~88%.
* fix(langflow): require litellm_params in A2A config instead of silent empty fallback
* fix(langflow): scope A2A session_id to the authenticated key
The LangFlow A2A bridge used the LangFlow session_id verbatim from the
client-controlled A2A contextId, so two distinct virtual keys authorized for
the same agent could read or append to each other's LangFlow conversation
memory by reusing a contextId.
Hand the authenticated key hash to the completion bridge through litellm_params
and namespace the forwarded session_id with it. The same key keeps a stable
session across turns, while different keys can no longer collide on a shared
contextId. The principal is hashed before it is embedded in the session_id, so
the stored token is never sent to the LangFlow backend; the original contextId
is preserved as a suffix for operator-side correlation.
* fix(langflow): wire authenticated key hash through A2A bridge and tests
Define A2A_USER_API_KEY_HASH_PARAM in the completion bridge handler, strip it
before litellm.acompletion, inject the authenticated key hash at the proxy A2A
endpoint, and add regression tests for per-key LangFlow session scoping.
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(tests): drop module-level test calls that break local_testing collection
Several files in tests/local_testing invoked their test functions at module
scope (e.g. test_register_model.py ran test_update_model_cost_via_completion()
at the bottom of the file). Those calls execute during pytest collection, so
they fire real network requests at import time. test_register_model.py's call
hit an OpenAI 429 and raised, turning into a collection error.
A collection error aborts the whole session for every job that globs
tests/local_testing/**/test_*.py, which is why unrelated jobs like
langfuse_logging_unit_tests (-k langfuse) and litellm_assistants_api_testing
(-k assistants) both failed even though neither touches register_model;
the -k filter only applies after collection.
pytest discovers and runs these test_* functions on its own, so the top-level
calls were dead and harmful. Removes them from test_register_model.py,
test_wandb.py, test_lunary.py, and test_multiple_deployments.py, and adds a
regression test that scans the directory for module-level test invocations.
* test(local_testing): skip unparseable files in module-scope invocation guardrail
A syntax error in any tests/local_testing file would make ast.parse raise an
unhandled SyntaxError, so the guardrail itself would crash with a confusing
traceback instead of its assertion message. Such a file already fails pytest
collection on its own, which is the clearer signal, so the guardrail now skips
files it cannot parse and stays focused on detecting module-scope test calls.
Reads files as utf-8 for deterministic behavior across platforms.
* feat(proxy): add per-MCP-server RPM rate limiting for keys and teams
Adds mcp_rpm_limit, a dict keyed by MCP server name (alias if set, else the
configured name) that caps requests per minute per server for a key or team.
The v3 rate limiter builds a per-server descriptor only when a limit is
configured for the server being called, so other servers stay uncapped and no
TPM reservation is engaged. Server identity is surfaced into the request data
via mcp_rate_limit_server_name so the limiter can resolve it.
* fix(proxy): gate MCP rpm descriptors on call_mcp_tool; document mcp_rpm_limit param
Only honor mcp_server_name when the call is an actual MCP tool call. Without
this, a normal LLM request could inject mcp_server_name in its body to consume
a target server's MCP quota and 429 legitimate tool calls. Also adds the
mcp_rpm_limit parameter docstring to update_key, new_user, and user_update so
the API docs validator passes.
* Fix MCP rate limit quota handling
* Delete scripts/test_mcp_rpm_limit.sh
* docs(proxy): clarify mcp_rpm_limit is enforced for keys and teams, not per user
* fix(proxy): accept mcp_rpm_limit in generate_key_helper_fn
NewUserRequest and GenerateKeyRequest inherit mcp_rpm_limit from
GenerateRequestBase, so /user/new and /key/generate forwarded the field
to generate_key_helper_fn, which did not accept it and returned a 500
("unexpected keyword argument 'mcp_rpm_limit'"). Accept the param and
store it in metadata, matching model_rpm_limit/model_tpm_limit, so the
limit is persisted where get_key_mcp_rpm_limit reads it.
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
