fix: add missing OpenAI chat completion params to OPENAI_CHAT_COMPLETION_PARAMS

[ghost]

What

Add 5 missing OpenAI chat completion parameters to OPENAI_CHAT_COMPLETION_PARAMS in litellm/constants.py:

• store
• prompt_cache_key
• prompt_cache_retention
• safety_identifier
• verbosity

Why

These params are defined in the OpenAI Chat Completions API and already present in DEFAULT_CHAT_COMPLETION_PARAM_VALUES, but were missing from OPENAI_CHAT_COMPLETION_PARAMS. This means they get dropped when filtering allowed params for OpenAI-compatible providers.

How

Added the 5 params to the OPENAI_CHAT_COMPLETION_PARAMS list.

Similar to #21195.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Feb 17, 2026 4:33am

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
5 out of 6 committers have signed the CLA.

✅ ishaan-jaff
✅ yuneng-jiang
✅ Chesars
✅ Sameerlite
✅ krrishdholakia
❌ shin-bot-litellm
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[greptile-apps]

Greptile Summary

Adds 5 missing OpenAI chat completion parameters (store, prompt_cache_key, prompt_cache_retention, safety_identifier, verbosity) to the OPENAI_CHAT_COMPLETION_PARAMS list in litellm/constants.py. These parameters were already defined in DEFAULT_CHAT_COMPLETION_PARAM_VALUES but were missing from OPENAI_CHAT_COMPLETION_PARAMS, causing them to be silently dropped when filtering allowed params for OpenAI-compatible providers (via get_standard_openai_params in litellm/utils.py, LiteLLMProxyChatConfig, HumanloopPromptManager, and the speech-to-completion bridge).

• Adds store, prompt_cache_key, prompt_cache_retention, safety_identifier, and verbosity to OPENAI_CHAT_COMPLETION_PARAMS
• No tests were added to verify the fix. The PR template requires at least 1 test in tests/litellm/.

Confidence Score: 4/5

• This PR is safe to merge — it only appends entries to a constants list, aligning it with an already-defined sibling dict.
• The change is minimal, additive, and aligns OPENAI_CHAT_COMPLETION_PARAMS with DEFAULT_CHAT_COMPLETION_PARAM_VALUES. All 5 params are well-established in the codebase. No behavioral regressions are expected. Deducted 1 point for the absence of a test verifying the fix.
• No files require special attention.

Important Files Changed

Filename	Overview
litellm/constants.py	Adds 5 missing OpenAI chat completion params (store, prompt_cache_key, prompt_cache_retention, safety_identifier, verbosity) to OPENAI_CHAT_COMPLETION_PARAMS. All params are already present in DEFAULT_CHAT_COMPLETION_PARAM_VALUES. No issues with the additions themselves.

Flowchart

flowchart TD
    A["Client sends chat completion request\nwith params like store, verbosity, etc."] --> B["litellm.completion()"]
    B --> C{"Filter params using\nOPENAI_CHAT_COMPLETION_PARAMS"}
    C -->|"Before fix:\nparams dropped"| D["Params missing from list\n→ silently removed"]
    C -->|"After fix:\nparams retained"| E["Params in list\n→ passed to provider"]
    E --> F["get_standard_openai_params()"]
    E --> G["LiteLLMProxyChatConfig\n.get_supported_openai_params()"]
    E --> H["HumanloopPromptManager"]
    E --> I["SpeechToCompletionBridge"]
    F --> J["OpenAI API / Compatible Provider"]
    G --> J
    H --> J
    I --> J

Loading

_{Last reviewed commit: 4ca2cef}

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}