[Bug]: gemini/gemma-3-27b-it function calling is not enabled exception

[theimpostor]

What happened?

model json shows that gemini/gemma-3-27b-it supports function calling, however trying it out via the attached script reports the exception below. Using openai-agents v0.0.13 which I believe is using LiteLLM 1.67.2.

repro script:

#!/usr/bin/env -S uv run --script

# /// script
# requires-python = ">=3.13"
# dependencies = [
#     "openai-agents[litellm]",
# ]
# ///

import asyncio

from agents import Agent, Runner, function_tool, set_tracing_disabled
from agents.extensions.models.litellm_model import LitellmModel

set_tracing_disabled(disabled=True)

@function_tool
def get_weather(city: str):
    print(f"[debug] getting weather for {city}")
    return f"The weather in {city} is sunny."


async def main(model: str, api_key: str):
    agent = Agent(
        name="Assistant",
        instructions="You only respond in haikus.",
        model=LitellmModel(model=model, api_key=api_key),
        tools=[get_weather],
    )

    result = await Runner.run(agent, "What's the weather in Tokyo?")
    print(result.final_output)


if __name__ == "__main__":
    # First try to get model/api key from args
    import argparse

    parser = argparse.ArgumentParser()
    parser.add_argument("--model", type=str, required=False)
    parser.add_argument("--api-key", type=str, required=False)
    args = parser.parse_args()

    model = args.model
    if not model:
        model = input("Enter a model name for Litellm: ")

    api_key = args.api_key
    if not api_key:
        api_key = input("Enter an API key for Litellm: ")

    asyncio.run(main(model, api_key))

Relevant log output

❯ ./litellm-tool-sample.py --api-key="..." --model=gemini/gemma-3-27b-it
Installed 56 packages in 375ms

Provider List: https://docs.litellm.ai/docs/providers


Give Feedback / Get Help: https://github.com/BerriAI/litellm/issues/new
LiteLLM.Info: If you need to debug this error, use `litellm._turn_on_debug()'.

Traceback (most recent call last):
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py", line 1285, in async_completion
    response = await client.post(
               ^^^^^^^^^^^^^^^^^^
        api_base, headers=headers, json=cast(dict, request_body)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )  # type: ignore
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/litellm_core_utils/logging_utils.py", line 135, in async_wrapper
    result = await func(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/llms/custom_httpx/http_handler.py", line 256, in post
    raise e
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/llms/custom_httpx/http_handler.py", line 212, in post
    response.raise_for_status()
    ~~~~~~~~~~~~~~~~~~~~~~~~~^^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/httpx/_models.py", line 829, in raise_for_status
    raise HTTPStatusError(message, request=request, response=self)
httpx.HTTPStatusError: Client error '400 Bad Request' for url 'https://generativelanguage.googleapis.com/v1beta/models/gemma-3-27b-it:generateContent?key=AIzaSyAIYQqE4rP2Vizu2GDND1qG8a-WaPwNlkw'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/main.py", line 477, in acompletion
    response = await init_response
               ^^^^^^^^^^^^^^^^^^^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py", line 1291, in async_completion
    raise VertexAIError(
    ...<3 lines>...
    )
litellm.llms.vertex_ai.common_utils.VertexAIError: {
  "error": {
    "code": 400,
    "message": "Function calling is not enabled for models/gemma-3-27b-it",
    "status": "INVALID_ARGUMENT"
  }
}


During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/private/var/folders/ct/x2gct7yn2bxfqs891n8h1dxr0000gn/T/tmp.i86mChDqDW/./litellm-tool-sample.py", line 52, in <module>
    asyncio.run(main(model, api_key))
    ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/Cellar/python@3.13/3.13.3/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ~~~~~~~~~~^^^^^^
  File "/usr/local/Cellar/python@3.13/3.13.3/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^
  File "/usr/local/Cellar/python@3.13/3.13.3/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/base_events.py", line 719, in run_until_complete
    return future.result()
           ~~~~~~~~~~~~~^^
  File "/private/var/folders/ct/x2gct7yn2bxfqs891n8h1dxr0000gn/T/tmp.i86mChDqDW/./litellm-tool-sample.py", line 31, in main
    result = await Runner.run(agent, "What's the weather in Tokyo?")
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/agents/run.py", line 218, in run
    input_guardrail_results, turn_result = await asyncio.gather(
                                           ^^^^^^^^^^^^^^^^^^^^^
    ...<19 lines>...
    )
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/agents/run.py", line 757, in _run_single_turn
    new_response = await cls._get_new_response(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<10 lines>...
    )
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/agents/run.py", line 916, in _get_new_response
    new_response = await model.get_response(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<10 lines>...
    )
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/agents/extensions/models/litellm_model.py", line 81, in get_response
    response = await self._fetch_response(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<9 lines>...
    )
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/agents/extensions/models/litellm_model.py", line 273, in _fetch_response
    ret = await litellm.acompletion(
          ^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<18 lines>...
    )
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/utils.py", line 1460, in wrapper_async
    raise e
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/utils.py", line 1321, in wrapper_async
    result = await original_function(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/main.py", line 496, in acompletion
    raise exception_type(
          ~~~~~~~~~~~~~~^
        model=model,
        ^^^^^^^^^^^^
    ...<3 lines>...
        extra_kwargs=kwargs,
        ^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 2214, in exception_type
    raise e
  File "/Users/shoda/.cache/uv/environments-v2/litellm-tool-sample-09def778d79087ab/lib/python3.13/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 1217, in exception_type
    raise BadRequestError(
    ...<11 lines>...
    )
litellm.exceptions.BadRequestError: litellm.BadRequestError: VertexAIException BadRequestError - {
  "error": {
    "code": 400,
    "message": "Function calling is not enabled for models/gemma-3-27b-it",
    "status": "INVALID_ARGUMENT"
  }
}

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

v1.67.2

Twitter / LinkedIn details

No response

[nileshtrivedi]

It should be possible to mimick tool-calling by adding preprocessing for the prompt and post-processing the response. Perhaps that's not in the scope of Litellm and should be built outside.

[zbloss]

I am seeing the same issue. Gemma3 supports function calling but LiteLLM throws an error saying it does not.

[nileshtrivedi]

People running into this issue might find this solution helpful for making Gemma3 on Ollama toolcalling-compatible: https://github.com/IllFil/gemma3-ollama-tools