litellm/tests
mika-lt 4212f1af27
fix(passthrough): forward OpenAI-Beta and default agents=v1 on the OpenAI passthrough (#49009)
* fix(passthrough): forward OpenAI-Beta and default agents=v1 on the OpenAI passthrough

The OpenAI-shaped passthrough now forwards the client's own OpenAI-Beta header (agents, chatkit, realtime and any future beta), keeps an operator's extra header over it, and injects agents=v1 on /v1/agents and /v1/vaults paths when the client sent none. Unbilled Agents API calls now log the agent's model on the spend row instead of unknown.

* fix(passthrough): keep OpenAI-Beta forwarding on OpenAI only and name the model on streamed and saved agent rows

Azure and Azure AI keep main's assistants=v2 rule. ChatKit paths get no route default. Streamed agents rows read session.agent.model from the session events instead of defaulting to gpt-4o, and saved agents name their top-level model.

* fix(passthrough): forward the client's OpenAI-Beta only on agents, vaults, and ChatKit paths

---------

Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
2026-10-29 05:52:34 -07:00
..
_support
audio_tests
base_sdk_tests refactor(rust-bridge): require the native extension (#48852) 2026-10-28 13:16:56 -07:00
benchmarks test(json_fragment_accumulator): assert peak allocation instead of a drain timing ratio (#43370) 2026-10-28 22:04:47 -07:00
code_coverage_tests feat(responses): support Codex client tool_search on Anthropic and other bridged providers (#48972) 2026-10-29 00:29:40 -07:00
documentation_tests
e2e feat(responses): support Codex client tool_search on Anthropic and other bridged providers (#48972) 2026-10-29 00:29:40 -07:00
e2e_harness
fixtures
harness_e2e
integration fix(caching): read the cache key's model group from the router's metadata for the call (#48917) 2026-10-29 05:35:23 -07:00
rust-python-harness
test_litellm_rust fix(anthropic): drop the non-spec custom object from /v1/messages tool definitions (#48893) 2026-10-29 03:36:20 -07:00
typing
unit fix(passthrough): forward OpenAI-Beta and default agents=v1 on the OpenAI passthrough (#49009) 2026-10-29 05:52:34 -07:00
windows_tests refactor(rust-bridge): require the native extension (#48852) 2026-10-28 13:16:56 -07:00
__init__.py
_fake_openai_endpoint_server.py
_flush_vcr_cache.py
_live_test_helpers.py
_master_key.py
_openai_record_replay_proxy.py
_process_helpers.py
_vcr_conftest_common.py
_vcr_redis_persister.py
_wait_helpers.py
_ws_vcr.py
AGENTS.md
capturing_transport.py
eval_swe_bench.py
fake_openai_endpoint.py
gettysburg.wav
large_text.py
openai_batch_completions.jsonl
pyrightconfig.json
README.MD
test_logging.conf
test_rust_python_harness.py
VCR_TESTING.md
white_100x100.png

In total litellm runs 1000+ tests

[02/20/2025] Update:

To make it easier to contribute and map what behavior is tested,

we've started mapping the litellm directory in tests/unit

This folder can only run mock tests.