mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-26 01:12:21 +00:00
b9a180cde9
Follow-up to #28113. Same root cause — Anthropic's Messages API rejects `temperature` for Claude Opus 4.7, accepting only `top_p` — applied to the Databricks adapter. Reported by @Kontinuation in the original issue thread (#26444), with a detailed reproduction by @dgomez04 in the same thread. Mirrors the #28113 pattern: - `DatabricksConfig.get_supported_openai_params` filters out `temperature` / `top_p` when the model id matches the `claude-opus-4-7` family - `map_openai_params` adds the same defense-in-depth guard so a raw kwargs leak is caught at the second layer too - `supports_temperature: false` / `supports_top_p: false` set on the new `databricks/databricks-claude-opus-4-7` entry in both `model_prices_and_context_window.json` and the backup file to satisfy `ci_cd/check_files_match.py` Helper is kept self-contained on `DatabricksConfig` (rather than reaching into the helper added by #28113 on `AnthropicConfig`) so this PR can land independently of #28113. Happy to deduplicate in a follow-up once both land. Tests: - new unit: temperature filtered out for databricks-claude-opus-4-7, the dated variant, and the vendor-prefixed form - new unit: unreleased dated 4.7 snapshots fall back to the `_is_claude_4_7_model` family check - regression: Sonnet 4.5 / Haiku 4.5 / Opus 4.5 / Opus 4.1 / 3.7-sonnet on Databricks still expose temperature - defense-in-depth: `map_openai_params(drop_params=False)` strips temperature for Opus 4.7 and preserves it for Opus 4.5 - end-to-end shape: top-level `litellm.get_supported_openai_params` for `databricks-claude-opus-4-7` no longer reports temperature (the exact contract `drop_params=True` keys off) Closes the Databricks gap from #26444. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| agent_tests | ||
| audio_tests | ||
| basic_proxy_startup_tests | ||
| batches_tests | ||
| benchmarks | ||
| code_coverage_tests | ||
| documentation_tests | ||
| enterprise | ||
| guardrails_tests | ||
| image_gen_tests | ||
| litellm | ||
| litellm-proxy-extras | ||
| litellm_core_utils | ||
| litellm_utils_tests | ||
| llm_responses_api_testing | ||
| llm_translation | ||
| load_tests | ||
| local_testing | ||
| logging_callback_tests | ||
| mcp_tests | ||
| multi_instance_e2e_tests | ||
| ocr_tests | ||
| old_proxy_tests/tests | ||
| openai_endpoints_tests | ||
| otel_tests | ||
| pass_through_tests | ||
| pass_through_unit_tests | ||
| proxy_admin_ui_tests | ||
| proxy_e2e_anthropic_messages_tests | ||
| proxy_security_tests | ||
| proxy_unit_tests | ||
| router_unit_tests | ||
| scim_tests | ||
| search_tests | ||
| spend_tracking_tests | ||
| store_model_in_db_tests | ||
| test_litellm | ||
| unified_google_tests | ||
| vector_store_tests | ||
| windows_tests | ||
| __init__.py | ||
| _flush_vcr_cache.py | ||
| _vcr_conftest_common.py | ||
| _vcr_redis_persister.py | ||
| eval_swe_bench.py | ||
| gettysburg.wav | ||
| large_text.py | ||
| openai_batch_completions.jsonl | ||
| README.MD | ||
| test_budget_management.py | ||
| test_callbacks_on_proxy.py | ||
| test_config.py | ||
| test_debug_warning.py | ||
| test_default_encoding_non_root.py | ||
| test_end_users.py | ||
| test_entrypoint.py | ||
| test_fallbacks.py | ||
| test_gpt5_azure_temperature_support.py | ||
| test_health.py | ||
| test_keys.py | ||
| test_litellm_proxy_responses_config.py | ||
| test_logging.conf | ||
| test_models.py | ||
| test_new_vector_store_endpoints.py | ||
| test_openai_endpoints.py | ||
| test_organizations.py | ||
| test_otel_thread_leak.py | ||
| test_passthrough_endpoints.py | ||
| test_presidio_latency.py | ||
| test_proxy_server_non_root.py | ||
| test_ratelimit.py | ||
| test_resource_cleanup.py | ||
| test_service_logger_otel.py | ||
| test_spend_logs.py | ||
| test_team.py | ||
| test_team_logging.py | ||
| test_team_members.py | ||
| test_users.py | ||
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/test_litellm
This folder can only run mock tests.