Description

#82 asks for a way to set the LLM call timeout when large inputs time out. Gemini exposes http_options/retry settings and Ollama has a timeout argument, but OpenAILanguageModel has no equivalent: timeout and max_retries passed via language_model_params land in _extra_kwargs, are dropped by the request-parameter allowlist in infer(), and never reach openai.OpenAI(...).

This adds timeout and max_retries as keyword arguments on OpenAILanguageModel.__init__ and forwards them to the openai.OpenAI client only when set, so SDK defaults are unchanged otherwise. They are placed after max_workers and before **kwargs, so existing positional calls are unaffected. This also applies to OpenAI-compatible endpoints configured with base_url.

Usage:

lx.extract(..., model_id="gpt-4o-mini",
           language_model_params={"timeout": 300, "max_retries": 5})

Verified through factory.create_model with the real SDK: timeout=7.5, max_retries=0 yields a client with those values; without them the client keeps the SDK defaults (Timeout(connect=5.0, read=600, ...), max_retries=2).

Overlap note: #548 also touches the openai.OpenAI(...) construction (adds default_headers); the two should merge with a trivial rebase.

Fixes #82

Feature

How Has This Been Tested?

Added tests that the client receives timeout/max_retries when set and is constructed exactly as before when they are not.

$ pytest tests --deselect tests/create_provider_plugin_test.py
908 passed, 21 skipped
$ isort langextract tests && pyink langextract tests --config pyproject.toml
$ pylint --rcfile=.pylintrc langextract/providers/openai.py

tests/create_provider_plugin_test.py has 3 failures on main in my environment (offline editable install / build isolation) unrelated to this change.

Checklist:

• I have read and acknowledged Google's Open Source Code of conduct.
• I have read the Contributing page, and I either signed the Google Individual CLA or am covered by my company's Corporate CLA.
• I have discussed my proposed solution with code owners in the linked issue(s) and we have agreed upon the general approach.
• I have made any needed documentation changes, or noted in the linked issue(s) that documentation elsewhere needs updating.
• I have added tests, or I have ensured existing tests cover the changes
• I have followed Google's Python Style Guide and ran pylint over the affected code.