Initial Checks

• I'm using the latest version of Pydantic AI
• I've searched for my issue in the issue tracker before opening this issue

Description

OpenRouterReasoning.effort is missing three valid values: xhigh, minimal, and none. The current type is Literal['high', 'medium', 'low'], but the OpenRouter API docs show six valid values: xhigh, high, medium, low, minimal, and none.

This isn't a critical bug as the code runs fine at runtime, only throwing errors for typecheckers.

Minimal, Reproducible Example

from pydantic_ai.models.openrouter import OpenRouterModelSettings

# All three produce pyright errors:
#   Type "Literal['xhigh']" is not assignable to type "Literal['high', 'medium', 'low']"
settings: OpenRouterModelSettings = {'openrouter_reasoning': {'effort': 'xhigh'}}

settings2: OpenRouterModelSettings = {'openrouter_reasoning': {'effort': 'minimal'}}

settings3: OpenRouterModelSettings = {'openrouter_reasoning': {'effort': 'none'}}

Logfire Trace

No response

Python, Pydantic AI & LLM client version

• Python: 3.13
• Pydantic AI: 1.69.0
• LLM provider SDK: n/a