How to make a LiteLLM models run in Reasoning mode

[LuSrodri]

How to make the LiteLLM models run in Reasoning mode?
I saw that exists the reasoning prop, but it doesn't clear to use...

model = LitellmModel(model="gemini/gemini-2.5-pro-preview-05-06", api_key=GEMINI_API_KEY)

agent= Agent[Context](
        name="An Agent",
        model=gemini_model,
        instructions=f"""
           Useful Agent
        """,
        model_settings=ModelSettings(tool_choice="auto", reasoning="????")
    )

[ukayani]

from agents import Agent, ModelSettings, Runner
from openai.types.shared.reasoning import Reasoning
from agents.extensions.models.litellm_model import LitellmModel


model = LitellmModel(model="anthropic/claude-3-7-sonnet-20250219")

agent = Agent(
    name="Assistant",
    instructions="You are a helpful assistant",
    model=model,
    model_settings=ModelSettings(reasoning=Reasoning(effort="high")),
)

result = Runner.run_sync(agent, "Write a haiku about recursion in programming.")

[LuSrodri]

Thanks! I had tried to found more about this Reasoning object type, but I didn`t have success :(

[ukayani]

unfortunately their abstraction does not allow controlling the reasoning budget, you only get 3 options. Anthropic for example allows budget of 16k but that isn't possible with "high".

[github-actions]

This issue is stale because it has been open for 7 days with no activity.

[github-actions]

This issue was closed because it has been inactive for 3 days since being marked as stale.