mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-28 06:23:07 +00:00
db6c70e9ee
Since the Add Auto Router form picks a deployed decision model (#45913), a provider has to be a Decisions provider to appear there. Nadir now is: `nadir/bucket` is a cost-map decision model, NadirDecisionsConfig serves it from POST /v1/bucket, and a Nadir deployment answers /v1/decisions, /v1/systemone, the health check and the auto router's Decision model. Nadir is a tier classifier, not a general judge, so it answers exactly one choice question whose choices include simple, medium and complex (matched case-insensitively, answered on the question's own keys) and refuses any other question set with a 400 before a call. A state shaped {"messages": [...]} or OpenAI-format input goes out as the conversation, any other state as the prompt. When the auto router's decision deployment is Nadir it sends the budgeted, role-tagged conversation as that state rather than the LLM-judge payload, the same input `provider: nadir` already gets. The `provider: nadir` path now uses the same config for its URL, key scoping and verdict parsing, so litellm/llms/nadir/common_utils.py is gone. NADIR_API_KEY still only goes to NADIR_API_BASE or the hosted API. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| _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.