Gemini OpenAI Compatibility: Multiple Functions Support in Function Calling - Error 400

Gemini API
1

Hi everyone,

I’m trying to implement function calling using Gemini’s OpenAI compatibility layer with gemini-1.5-flash-002. While the basic example with a single function works perfectly, I’m getting an error when trying to add multiple functions to the tools array.

Here’s my code:

tools = [
  {
    "type": "function",
    "function": {
      "name": "get_weather",
      "description": "Get the weather in a given location",
      "parameters": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string",
            "description": "The city and state, e.g. Chicago, IL",
          },
          "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
        },
        "required": ["location"],
      },
    }
  },
  {
    "type": "function",
    "function": {
      "name": "get_weather_forecast",
      "description": "Get the weather forecast for the next few days",
      # ... similar parameters plus 'days'
    }
  }
]

This results in:

openai.BadRequestError: Error code: 400 - [{‘error’: {‘code’: 400, ‘message’: ‘Request contains an invalid argument.’, ‘status’: ‘INVALID_ARGUMENT’}}]

The same code works fine with a single function. Is there a current limitation on the number of functions supported in the OpenAI compatibility mode? The documentation doesn’t mention this as a limitation, but I noticed the examples only show single function usage.
Any insights or workarounds would be appreciated!

1 Like
3

I have encountered the same issue. When I use the OpenAI API, the same tools do not cause any problems and work as expected. I suspect it is a compatibility issue between Gemini and the OpenAI standard.