litellm's default LiteLLMAiohttpTransport routes requests through aiohttp,
which sits below httpx and is invisible to vcrpy's httpx-stub interception.
Under vcrpy + aiohttp, requests reach the real network but responses come
back through the stubbed httpx transport as empty 200s, surfacing as
'Unable to get json response - Expecting value: line 1 column 1 (char 0)'
in providers like Anthropic, Gemini, and any other path that exercises the
aiohttp transport.
Disabling the aiohttp transport when the VCR persister is registered
forces all calls through pure httpx, which vcrpy can record and replay
correctly.
record_mode='once' refused to add new requests once any cassette
existed in Redis. Combined with filter_non_2xx_response (which drops
non-2xx responses from the saved cassette) and a 24h shared-Redis TTL,
a single transient API failure mid-test left the cassette stuck with
only the leading non-API requests (e.g. the model_prices fetch from
raw.githubusercontent.com), and every subsequent run for the next 24h
errored with 'Can't overwrite existing cassette'.
new_episodes records anything not already present, so partially
populated cassettes recover on the next run instead of poisoning the
suite for a full TTL window.
Removes commentary that restated the code, including:
- module-level banners explaining what the conftest does (covered by
Readme.md and the function bodies)
- docstrings on _scrub_response, _before_record_response, vcr_config,
_vcr_disabled, pytest_recording_configure (function names + bodies
are self-evident)
- inline notes about header filtering, match_on, etc.
- per-test docstrings restating the test name
Keeps the two non-obvious notes that aren't recoverable from the code:
the vcrpy/respx httpx-transport collision rationale on
_RESPX_CONFLICTING_FILES, the vcrpy "return None to skip persisting"
contract on filter_non_2xx_response, and the fixture-ordering
dependency on _vcr_record_retries.
Removes the YAML cassette feature entirely and replaces it with a
Redis-only flow. Every test in tests/llm_translation/ and
tests/llm_responses_api_testing/ is auto-marked @pytest.mark.vcr via
conftest.pytest_collection_modifyitems, so any provider call lands in
the Redis cache (litellm:vcr:cassette:<rel_path>, 24h TTL). First run
records, runs within the day replay, day rollover re-records and
surfaces upstream API drift within 24h.
VCR is on by default. Set LITELLM_VCR_DISABLE=1, or simply leave
REDIS_HOST unset, to opt out — both bypass the auto-marker entirely so
nothing about cassettes runs. record_mode is "once" so cache-miss
records and cache-hit replays.
The 8 existing respx-using files in tests/llm_translation are excluded
from the auto-marker (vcrpy and respx both patch the httpx transport;
applying both makes one silently win). The persister's own unit-test
file is also excluded so it doesn't recursively run inside a cassette.
The persister moved from tests/llm_translation/_vcr_redis_persister.py
to tests/_vcr_redis_persister.py so both conftests share it. The two
demo tests in test_anthropic_completion_vcr.py were ported into
test_anthropic_completion.py and the demo file was deleted.
Adds tests/_flush_vcr_cache.py + a Make target
(test-llm-translation-flush-vcr-cache) that scans
litellm:vcr:cassette:* and pipelines DELETEs, for the
"I want the next CI run to re-record now" workflow. Drops the now-dead
test-llm-translation-record target.
Provider keys are still required on cache-miss (which happens on first
run and once a day after that). Replay-mode runs need only Redis.
* fix(responses): fix O(n²) CPU overhead in reasoning streaming path
stream_chunk_builder was called on every reasoning chunk, rebuilding the
entire response from all collected chunks each time. Replace with
incremental accumulation of reasoning_content parts, only joining at
reasoning end.
* fix(responses): eliminate per-chunk thread spawning in async streaming path
_process_chunk() called run_async_function() on every SSE chunk, which
when invoked from an async context spawns a thread + event loop per call.
Move the hook call out of _process_chunk into the callers: async __anext__
directly awaits it, sync __next__ uses run_async_function.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* perf: reduce responses streaming CPU for text-only streams
* fix(test): replace deprecated claude-3-7-sonnet-latest in responses API test
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(test): replace deprecated claude-3-7-sonnet-latest in tool result fix test
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(test): replace deprecated claude-3-7-sonnet-latest in tool result empty call_id test
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Replace copy.deepcopy with model_dump + model_validate in streaming
iterator logging to handle Pydantic ValidatorIterator objects that
cannot be pickled when tool_choice uses allowed_tools mode.
Co-authored-by: Krish Dholakia <krrishdholakia@gmail.com>
* Add openai metadata filed in the request
* Add docs related to openai metadata
* Add utils
* test_completion_openai_metadata[True]
* Added support for though signature for gemini 3 in responses api (#16872)
* Added support for though signature for gemini 3
* Update docs with all supported endpoints and cost tracking
* Added config based routing support for batches and files
* fix lint errors
* Litellm anthropic image url support (#16868)
* Add image as url support to anthropic
* fix mypy errors
* fix tests
* Fix: Populate spend_logs_metadata in batch and files endpoints (#16921)
* Add spend-logs-metadata to the metadata
* Add tests for spend logs metadata in batches
* use better names
* Remove support for penalty param for gemini 3 (#16907)
* Remove support for penalty param
* remove halucinated model names
* fix mypy/test errors
* fix tests
* fix too many lines error
* fix too many lines error
* Add config for cicd test case
* Fix final tests
* fix batch tests
* fix batch tests
## Problem
The `extra_body` parameter in `litellm.responses()` and `litellm.aresponses()`
was being accepted but never passed to the HTTP request sent to the LLM provider.
This prevented users from sending custom/experimental parameters to provider APIs.
## Changes
- Added `data.update(extra_body)` in `async_response_api_handler` (line 2138)
- Added `data.update(extra_body)` in `response_api_handler` (line 2012)
- Added tests to `test_openai_responses_api.py` for extra_body functionality
## Testing
- Tests verify extra_body params are passed in both sync and async modes
- Existing Responses API tests continue to pass
- Manually verified with OpenAI API that custom params are sent correctly
## Impact
Users can now pass custom/experimental parameters via extra_body:
```python
litellm.aresponses(
model="gpt-4o",
input="hello",
extra_body={"custom_param": "value"} # Now works!
)
```
This aligns with the OpenAI SDK pattern and matches behavior in other
LiteLLM endpoints (completion, embedding, etc.) that already support extra_body.
This commit fixes two bugs in Responses API streaming tests:
1. **Usage field naming bug**: Tests were using `input_tokens` and
`output_tokens` but the Usage object uses `prompt_tokens` and
`completion_tokens`.
2. **Missing cost in streaming usage**: When `include_cost_in_streaming_usage`
was enabled, the cost was calculated and added to ResponseAPIUsage, but was
lost during the transformation to the Usage object.
Changes:
- Updated test assertions to use correct field names (prompt_tokens, completion_tokens)
- Added cost preservation logic in FakeStreamerResponsesAPIIterator
- Modified _transform_response_api_usage_to_chat_usage() to preserve cost attribute
All streaming tests now pass successfully.
