Fix: add stop param as supported for openai and azure#21539
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Greptile SummaryThis PR attempts to fix parameter handling for the Key Issues
What Changed
Architecture ConcernThe approach creates conflicting parameter handling logic. The global Confidence Score: 1/5
|
| Filename | Overview |
|---|---|
| litellm/constants.py | Removed store from OPENAI_CHAT_COMPLETION_PARAMS - this could be problematic as it may break functionality added in PR #21195 |
| litellm/llms/azure/chat/gpt_transformation.py | Added store to Azure's supported params list and removed trailing whitespace |
| litellm/llms/openai/chat/gpt_transformation.py | Added store to OpenAI's supported params list |
| tests/llm_translation/test_optional_params.py | Added comprehensive test for store parameter passing through OpenAI and Azure providers |
Last reviewed commit: 44a68cf
| "thinking", | ||
| "web_search_options", | ||
| "service_tier", | ||
| "store", | ||
| "prompt_cache_key", |
There was a problem hiding this comment.
Removing store from OPENAI_CHAT_COMPLETION_PARAMS will break the existing test test_store_in_openai_chat_completion_params() added in PR #21195 (litellm/constants.py:581). This test explicitly checks assert "store" in OPENAI_CHAT_COMPLETION_PARAMS and will fail.
Additionally, get_standard_openai_params() relies on this list to identify standard OpenAI parameters. Removing store means it will be filtered out when calling OpenAI/Azure, breaking the functionality that PR #21195 was supposed to fix.
Relevant issues
Pre-Submission checklist
Please complete all items before asking a LiteLLM maintainer to review your PR
tests/litellm/directory, Adding at least 1 test is a hard requirement - see detailsmake test-unit@greptileaiand received a Confidence Score of at least 4/5 before requesting a maintainer reviewCI (LiteLLM team)
Branch creation CI run
Link:
CI run for the last commit
Link:
Merge / cherry-pick CI run
Links:
Type
🐛 Bug Fix
Changes
config: