litellm/tests/router_unit_tests/test_router_index_management.py
Mateo Wang 2c733c00f5
chore(ci): modernize model references in tests and configs (#27856)
* test: modernize models used in CircleCI e2e test suites

Replaces obsolete models (gpt-4o, gpt-4o-mini, gpt-3.5-turbo,
claude-3-5-sonnet-20240620, claude-sonnet-4-20250514) with current
equivalents across the e2e_openai_endpoints and
proxy_e2e_anthropic_messages_tests CircleCI jobs.

- gpt-4o -> gpt-5.5 (responses API e2e tests)
- gpt-4o-mini -> gpt-5-mini (websocket responses, oai_misc_config)
- gpt-4o-mini-2024-07-18 -> gpt-4.1-mini-2025-04-14 (fine-tuning,
  still actively fine-tunable)
- gpt-4 / gpt-3.5-turbo target_model_names example -> gpt-5.5 /
  gpt-5-mini
- bedrock claude-3-5-sonnet-20240620 batch entry -> haiku-4-5-20251001
  (also aligning oai_misc_config model_name with what
  test_bedrock_batches_api.py actually requests)
- bedrock claude-sonnet-4-20250514 (deprecated, retires 2026-06-15)
  -> claude-sonnet-4-5-20250929

* test: point bedrock-claude-sonnet-4 alias at Sonnet 4.6, not 4.5

Greptile/Cursor flagged that after the previous commit, the
bedrock-claude-sonnet-4 alias collided with bedrock-claude-sonnet-4.5
(both pointed to claude-sonnet-4-5-20250929). Rename to
bedrock-claude-sonnet-4.6 and point it at the Sonnet 4.6 Bedrock ID
(us.anthropic.claude-sonnet-4-6, already in the litellm model
registry) so the alias name matches the underlying model version.

* test: modernize models across remaining CI-mounted configs & tests

Expands the modernization sweep to all CircleCI-mounted proxy configs
and to test directories where the model literal is a fixture/route key
(not the test's subject).

Config changes:
- proxy_server_config.yaml: bump gpt-3.5-turbo / gpt-3.5-turbo-1106 /
  gpt-4o / gemini-1.5-flash / dall-e-3 underlying models; rename
  gpt-3.5-turbo-end-user-test alias to gpt-5-mini-end-user-test; bump
  text-embedding-ada-002 underlying to text-embedding-3-small. User-
  facing aliases (gpt-3.5-turbo, gpt-4, text-embedding-ada-002, etc.)
  preserved for backward compatibility with tests.
- simple_config.yaml, otel_test_config.yaml, spend_tracking_config.yaml:
  bump gpt-3.5-turbo underlying to gpt-5-mini.
- pass_through_config.yaml: claude-3-5-sonnet / claude-3-7-sonnet /
  claude-3-haiku entries replaced with claude-sonnet-4-5 / claude-
  haiku-4-5 / claude-opus-4-7.
- oai_misc_config.yaml: align alias name with the gpt-5-mini rename.

Test changes (proactive: claude-sonnet-4-20250514 / claude-opus-4-
20250514 retire 2026-06-15):
- tests/llm_translation/test_anthropic_completion.py: bump 3 references
  + paired Vertex AI ID to claude-sonnet-4-5.
- tests/llm_translation/test_optional_params.py: bump 2 references.
- tests/pass_through_unit_tests/test_anthropic_messages_passthrough.py
  and test_bedrock_anthropic_messages_test.py: bump router fixtures
  using the deprecated model IDs.
- tests/pass_through_unit_tests/base_anthropic_messages_tool_search_test.py:
  modernize docstring examples.
- tests/test_end_users.py: update references to renamed alias.

* test: modernize placeholder model literals in router_unit_tests

Mass replace_all on fixture/placeholder model literals across the
router_unit_tests/ suite (model name is a routing key / label, not the
test subject). Sub-agent sweep so far — additional commits will follow
for logging_callback_tests/, enterprise/, top-level tests/test_*.py,
and other CI-mounted dirs.

Mappings applied:
- gpt-3.5-turbo -> gpt-5-mini
- gpt-4 (bare) -> gpt-5.5
- gpt-4o (bare) -> gpt-5
- text-embedding-ada-002 -> text-embedding-3-small
- claude-3-sonnet-20240229 / claude-3-opus-20240229 /
  claude-3-haiku-20240307 / claude-3-5-sonnet-20240620 ->
  claude-sonnet-4-5-20250929 / claude-opus-4-7 /
  claude-haiku-4-5-20251001 as appropriate

Explicitly preserved:
- gpt-4o-mini-* variants (transcribe, tts, etc.) where they're current
- gpt-4-turbo / gpt-4-vision-preview / gpt-4-0613 (subject literals)
- JSONL batch body literals
- Mock LLM response model fields (must match upstream)
- Fake/mock identifiers

* test: modernize placeholder model literals across remaining CI suites

Sub-agent sweep across logging_callback_tests/, guardrails_tests/,
enterprise/, pass_through_unit_tests/, otel_tests/,
llm_responses_api_testing/, batches_tests/, spend_tracking_tests/,
litellm_utils_tests/, unified_google_tests/, and a few top-level
tests/test_*.py files where the model literal is a fixture or
placeholder (router model_list, mock standard logging payload, mock
callback data) rather than the test's subject.

Mappings applied (see scope notes below):
- gpt-3.5-turbo -> gpt-5-mini
- gpt-4 (bare) -> gpt-5.5
- gpt-4o (bare) -> gpt-5.5 (corrected from initial gpt-5 — bare gpt-5
  is not a valid OpenAI alias; only gpt-5.5 / gpt-5.4 / gpt-5.2-codex
  / gpt-5-mini exist)
- gpt-4o-mini (bare) -> gpt-5-mini
- text-embedding-ada-002 -> text-embedding-3-small
- claude-3-sonnet-20240229 -> claude-sonnet-4-5-20250929
- claude-3-opus-20240229 -> claude-opus-4-7
- claude-3-haiku-20240307 -> claude-haiku-4-5-20251001
- claude-3-5-sonnet-20240620/20241022 -> claude-sonnet-4-5-20250929
- claude-3-7-sonnet-20250219 -> claude-sonnet-4-6
- gemini-1.5-flash -> gemini-2.5-flash
- gemini-1.5-pro -> gemini-2.5-pro

Explicitly preserved (not modernized):
- llm_translation/ tests where model is the SUBJECT (provider-specific
  translation/transformation logic). Only the deprecated 20250514
  references were already bumped in a prior commit.
- Cost-calc / tokenizer subject tests in test_utils.py (skip-ranges
  documented by the sub-agent).
- Bedrock model IDs in test_health_check.py path-stripping tests.
- JSONL batch request bodies and mock LLM response bodies (must match
  upstream literal).
- Langfuse expected-request-body JSON fixtures (cost values are exact-
  match-asserted; changing the model would shift response_cost).
- gpt-3.5-turbo-instruct (text-completion endpoint; no modern OpenAI
  equivalent).
- Top-level tests calling the proxy through user-facing aliases
  (gpt-3.5-turbo, gpt-4, text-embedding-ada-002, dall-e-3) — aliases
  in proxy_server_config.yaml stay; only the underlying model was
  bumped.
- tests/test_gpt5_azure_temperature_support.py (the test's whole point
  is model-name handling).
- Fake / mock / openai/fake identifiers.

Notable side fixes:
- test_spend_accuracy_tests.py: UPSTREAM_MODEL now matches what
  spend_tracking_config.yaml's proxy actually routes to (gpt-5-mini),
  resolving a latent inconsistency.
- proxy_server_config.yaml: bare `gpt-5` alias renamed to `gpt-5.5`
  (bare gpt-5 is not a valid OpenAI alias).
- test_batches_logging_unit_tests.py: explicit_models list entries
  kept distinct (gpt-5-mini + gpt-5.5) after bulk rename.

* test: fix CI failures from model modernization sweep

CI surfaced 4 categories of regression from the bulk modernization:

1. Azure deployment names are customer-specific. Reverted:
   - tests/litellm_utils_tests/test_health_check.py: azure/text-
     embedding-3-small -> azure/text-embedding-ada-002 (the CI Azure
     account does not have a text-embedding-3-small deployment).
   - tests/logging_callback_tests/test_custom_callback_router.py:
     same revert for two router fixtures driving aembedding.

2. gpt-5 family does not accept temperature != 1. Tests that pass a
   custom temperature swapped from gpt-5-mini to gpt-4.1-mini (modern
   non-reasoning OpenAI mini that still accepts temperature/logprobs):
   - tests/logging_callback_tests/test_datadog.py
   - tests/logging_callback_tests/test_langsmith_unit_test.py
   - tests/logging_callback_tests/test_otel_logging.py

3. proxy_server_config.yaml's gpt-3.5-turbo-large alias was routing to
   gpt-5.5 (a reasoning model that rejects logprobs). The proxy test
   tests/test_openai_endpoints.py::test_chat_completion_streaming
   exercises logprobs/top_logprobs through that alias. Bumped the
   underlying model to gpt-4.1 (non-reasoning, still modern).

4. tests/logging_callback_tests/test_gcs_pub_sub.py asserts against a
   pinned JSON fixture (gcs_pub_sub_body/spend_logs_payload.json) with
   hardcoded model="gpt-4o" and a model-specific spend value. Reverted
   the litellm.acompletion calls in the test to model="gpt-4o" so the
   fixture's exact-match assertions still hold.

5. tests/pass_through_unit_tests/test_anthropic_messages_passthrough.py:
   anthropic.messages.create routing to openai/gpt-5-mini returned an
   empty content[0] with max_tokens=100 (reasoning-token consumption).
   Swapped to openai/gpt-4.1-mini.

* test: fix Assistants API model + 2 cursor[bot] review nits

1. pass_through_unit_tests/test_custom_logger_passthrough.py: gpt-5.5
   isn't accepted by the /v1/assistants endpoint
   ("unsupported_model"). Switch to gpt-4.1-mini (modern, Assistants-
   API-supported, non-reasoning).

2. example_config_yaml/pass_through_config.yaml: the previous sweep
   bumped the claude-3-7-sonnet alias to claude-opus-4-7, which is a
   tier change (Sonnet -> Opus). Map to claude-sonnet-4-6 to keep the
   Sonnet tier intact. (Cursor bugbot review.)

3. example_config_yaml/simple_config.yaml: model_name was left as
   gpt-3.5-turbo while the underlying was bumped to gpt-5-mini, which
   muddles the "simple" example. Make both sides gpt-5-mini so the
   most basic example is a straight 1:1 mapping again. (Cursor bugbot
   review.)

* fix: revert gpt-4/gpt-3.5-turbo alias underlying to non-reasoning models

tests/test_openai_endpoints.py::test_completion calls the proxy alias
"gpt-4" with temperature=0, and other tests call gpt-3.5-turbo with
custom temperature / logprobs / the legacy /v1/completions endpoint.
The earlier modernization mapped both aliases to gpt-5.5 / gpt-5-mini,
which are reasoning models that reject temperature != 1 and don't
expose /v1/completions. Map the aliases to gpt-4.1 / gpt-4.1-mini
(modern non-reasoning OpenAI models) instead — keeps user-facing
aliases preserved while picking a current underlying that still
supports the parameters/endpoints the tests exercise.
2026-05-15 15:44:28 -07:00

331 lines
13 KiB
Python

import sys
import os
import pytest
import ast
import ast
sys.path.insert(
0, os.path.abspath("../..")
) # Adds the parent directory to the system path
from litellm import Router
class TestRouterIndexManagement:
"""Test cases for router index management functions"""
@pytest.fixture
def router(self):
"""Create a router instance for testing"""
return Router(model_list=[])
def test_deletion_updates_model_name_indices(self, router):
"""Test that deleting a deployment updates model_name_to_deployment_indices correctly"""
router.model_list = [
{"model_name": "gpt-3.5", "model_info": {"id": "model-1"}},
{"model_name": "gpt-5.5", "model_info": {"id": "model-2"}},
{"model_name": "gpt-5.5", "model_info": {"id": "model-3"}},
{"model_name": "claude", "model_info": {"id": "model-4"}},
]
router.model_id_to_deployment_index_map = {
"model-1": 0,
"model-2": 1,
"model-3": 2,
"model-4": 3,
}
router.model_name_to_deployment_indices = {
"gpt-3.5": [0],
"gpt-5.5": [1, 2],
"claude": [3],
}
# Remove one of the duplicate gpt-5.5 deployments
router._update_deployment_indices_after_removal(
model_id="model-2", removal_idx=1
)
# Verify indices are shifted correctly
assert router.model_name_to_deployment_indices["gpt-3.5"] == [0]
assert router.model_name_to_deployment_indices["gpt-5.5"] == [
1
] # was [1,2], removed 1, shifted 2->1
assert router.model_name_to_deployment_indices["claude"] == [
2
] # was [3], shifted to [2]
# Remove the last gpt-5.5 deployment
router._update_deployment_indices_after_removal(
model_id="model-3", removal_idx=1
)
# Verify gpt-5.5 is removed from dict when no deployments remain
assert "gpt-5.5" not in router.model_name_to_deployment_indices
assert router.model_name_to_deployment_indices["gpt-3.5"] == [0]
assert router.model_name_to_deployment_indices["claude"] == [1]
def test_build_model_id_to_deployment_index_map(self, router):
"""Test _build_model_id_to_deployment_index_map function"""
model_list = [
{
"model_name": "gpt-5-mini",
"litellm_params": {"model": "gpt-5-mini"},
"model_info": {"id": "model-1"},
},
{
"model_name": "gpt-5.5",
"litellm_params": {"model": "gpt-5.5"},
"model_info": {"id": "model-2"},
},
]
# Test: Build index from model list
router._build_model_id_to_deployment_index_map(model_list)
# Verify: model_list is populated
assert len(router.model_list) == 2
# Verify: model_id_to_deployment_index_map is correctly built
assert router.model_id_to_deployment_index_map["model-1"] == 0
assert router.model_id_to_deployment_index_map["model-2"] == 1
def test_add_model_to_list_and_index_map_from_model_info(self, router):
"""Test _add_model_to_list_and_index_map extracting model_id from model_info"""
# Setup: Empty router
router.model_list = []
router.model_id_to_deployment_index_map = {}
# Test: Add model without explicit model_id
model = {"model": "test-model", "model_info": {"id": "model-info-id"}}
router._add_model_to_list_and_index_map(model=model)
# Verify: Model added to list
assert len(router.model_list) == 1
assert router.model_list[0] == model
# Verify: Index map uses model_info.id
assert router.model_id_to_deployment_index_map["model-info-id"] == 0
def test_add_model_to_list_and_index_map_multiple_models(self, router):
"""Test _add_model_to_list_and_index_map with multiple models to verify indexing"""
# Setup: Empty router
router.model_list = []
router.model_id_to_deployment_index_map = {}
# Test: Add multiple models
model1 = {"model": "model1", "model_info": {"id": "id-1"}}
model2 = {"model": "model2", "model_info": {"id": "id-2"}}
model3 = {"model": "model3", "model_info": {"id": "id-3"}}
router._add_model_to_list_and_index_map(model=model1, model_id="id-1")
router._add_model_to_list_and_index_map(model=model2, model_id="id-2")
router._add_model_to_list_and_index_map(model=model3, model_id="id-3")
# Verify: All models added to list
assert len(router.model_list) == 3
assert router.model_list[0] == model1
assert router.model_list[1] == model2
assert router.model_list[2] == model3
# Verify: Correct indices in map
assert router.model_id_to_deployment_index_map["id-1"] == 0
assert router.model_id_to_deployment_index_map["id-2"] == 1
assert router.model_id_to_deployment_index_map["id-3"] == 2
def test_update_team_model_index(self, router):
"""Test _update_team_model_index updates team_model_to_deployment_indices."""
model = {
"model_name": "team-alias",
"model_info": {
"id": "dep-1",
"team_id": "team-abc",
"team_public_model_name": "gpt-5.5",
},
}
router._update_team_model_index(model, 0)
assert router.team_model_to_deployment_indices[("team-abc", "gpt-5.5")] == [0]
router._update_team_model_index(model, 2)
assert router.team_model_to_deployment_indices[("team-abc", "gpt-5.5")] == [0, 2]
router._update_team_model_index(
{"model_name": "x", "model_info": {"id": "dep-2"}}, 5
)
assert router.team_model_to_deployment_indices == {
("team-abc", "gpt-5.5"): [0, 2],
}
def test_has_model_id(self, router):
"""Test has_model_id function for O(1) membership check"""
# Setup: Add models to router
router.model_list = [
{"model": "test1", "model_info": {"id": "model-1"}},
{"model": "test2", "model_info": {"id": "model-2"}},
{"model": "test3", "model_info": {"id": "model-3"}},
]
router.model_id_to_deployment_index_map = {
"model-1": 0,
"model-2": 1,
"model-3": 2,
}
# Test: Check existing model IDs
assert router.has_model_id("model-1") == True
assert router.has_model_id("model-2") == True
assert router.has_model_id("model-3") == True
# Test: Check non-existing model IDs
assert router.has_model_id("non-existent") == False
assert router.has_model_id("") == False
assert router.has_model_id("model-4") == False
# Test: Empty router
empty_router = Router(model_list=[])
assert empty_router.has_model_id("any-id") == False
def test_build_model_name_index(self, router):
"""Test _build_model_name_index function"""
model_list = [
{
"model_name": "gpt-5-mini",
"litellm_params": {"model": "gpt-5-mini"},
"model_info": {"id": "model-1"},
},
{
"model_name": "gpt-5.5",
"litellm_params": {"model": "gpt-5.5"},
"model_info": {"id": "model-2"},
},
{
"model_name": "gpt-5.5", # Duplicate model_name, different deployment
"litellm_params": {"model": "gpt-5.5"},
"model_info": {"id": "model-3"},
},
]
# Test: Build index from model list
router._build_model_name_index(model_list)
# Verify: model_name_to_deployment_indices is correctly built
assert "gpt-5-mini" in router.model_name_to_deployment_indices
assert "gpt-5.5" in router.model_name_to_deployment_indices
# Verify: gpt-5-mini has single deployment
assert router.model_name_to_deployment_indices["gpt-5-mini"] == [0]
# Verify: gpt-5.5 has multiple deployments
assert router.model_name_to_deployment_indices["gpt-5.5"] == [1, 2]
# Test: Rebuild index (should clear and rebuild)
new_model_list = [
{
"model_name": "claude-3",
"litellm_params": {"model": "claude-3"},
"model_info": {"id": "model-4"},
},
]
router._build_model_name_index(new_model_list)
# Verify: Old entries are cleared
assert "gpt-5-mini" not in router.model_name_to_deployment_indices
assert "gpt-5.5" not in router.model_name_to_deployment_indices
# Verify: New entry is added
assert "claude-3" in router.model_name_to_deployment_indices
assert router.model_name_to_deployment_indices["claude-3"] == [0]
def test_no_linear_scans_in_router(self):
"""
Static analysis test to ensure Router doesn't use O(n) linear scans.
Scans router.py for 'in self.model_list' pattern which indicates
inefficient O(n) iteration instead of using index-based O(1) lookups.
Methods should use:
- model_id_to_deployment_index_map for O(1) model_id lookups
- model_name_to_deployment_indices for O(1) + O(k) model_name lookups
"""
# Methods that are allowed to iterate through self.model_list
ALLOWED_METHODS = [
"_get_deployment_by_litellm_model", # Edge case: lookup by litellm_params.model (not indexed)
"_finalize_adaptive_router_if_configured", # Init-time prefix scan for "auto_router/adaptive_router" (no index for prefix match)
]
# Get path to router.py
router_file = os.path.join(
os.path.dirname(os.path.dirname(os.path.dirname(__file__))),
"litellm",
"router.py",
)
# Read the file
with open(router_file, "r") as f:
content = f.read()
# Parse with AST
tree = ast.parse(content)
# Find violations
violations = []
ignore_methods = set(ALLOWED_METHODS)
for node in ast.walk(tree):
if isinstance(node, ast.FunctionDef):
method_name = node.name
# Skip ignored methods
if method_name in ignore_methods:
continue
# Get source for this method
try:
method_source = ast.get_source_segment(content, node)
if not method_source:
continue
# Check for the anti-pattern: "in self.model_list"
# This catches: for x in self.model_list, if x in self.model_list, etc.
if "in self.model_list" in method_source:
# Extract the specific line for better error reporting
lines = method_source.split("\n")
pattern_line = None
for line in lines:
if "in self.model_list" in line:
pattern_line = line.strip()
break
violations.append(
{
"method": method_name,
"line": node.lineno,
"pattern": pattern_line or "in self.model_list",
}
)
except Exception:
# Skip if we can't get source segment
pass
# Assert no violations
if violations:
error_msg = "\n".join(
[
f" - {v['method']}() at line {v['line']}: {v['pattern']}"
for v in violations
]
)
pytest.fail(
f"\n{'='*70}\n"
f"Found O(n) linear scan pattern in router.py:\n\n"
f"{error_msg}\n\n"
f"These methods should use index maps instead:\n"
f" - model_id_to_deployment_index_map (for model_id lookups)\n"
f" - model_name_to_deployment_indices (for model_name lookups)\n\n"
f"If a method legitimately needs O(n) iteration, add it to\n"
f"ALLOWED_METHODS in this test method.\n"
f"{'='*70}\n"
)
def test_model_names_is_set(self):
"""Verify that model_names uses a set for O(1) lookups, not a list (O(n))"""
router = Router(model_list=[])
assert isinstance(
router.model_names, set
), f"model_names should be a set for O(1) lookups, but got {type(router.model_names)}"