mirror of
https://github.com/BerriAI/litellm.git
synced 2026-11-05 07:37:50 +00:00
7d7066f495
Since #47419 a deployment with `litellm_params.model_id` and `model_info.base_model` is priced from the backend model's cost-map row through `model_info_with_pricing_overrides()`, which validates the merged row with `TypeAdapter(_PricingModelInfo)`. `ModelInfoBase.mode` is a closed Literal that has no `rerank`, so a Bedrock rerank deployment using an application inference profile failed validation while the router was built, the deployment was skipped, and every call to the model group returned "no healthy deployments". Add the cost-map modes that were missing from the Literal (`rerank`, `video_generation`, `image_edit`, `search`, `moderation`, `guardrail`, `vector_store`), and make the pricing merge non-fatal: if the row still does not validate, log a warning and keep the caller's explicit `model_info` instead of dropping the deployment. Fixes #48545 |
||
|---|---|---|
| .. | ||
| _support | ||
| audio_tests | ||
| base_sdk_tests | ||
| benchmarks | ||
| code_coverage_tests | ||
| documentation_tests | ||
| e2e | ||
| e2e_harness | ||
| fixtures | ||
| harness_e2e | ||
| integration | ||
| rust-python-harness | ||
| test_litellm_rust | ||
| typing | ||
| unit | ||
| windows_tests | ||
| __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.