Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.together.ai/llms.txt

Use this file to discover all available pages before exploring further.

Vision language models (VLMs) can also use function calling, letting you combine image understanding with tool use. This enables use cases like extracting structured data from images, identifying objects and taking actions, or analyzing visual content to trigger specific functions. 
import json
from together import Together

client = Together()

tools = [
    {
        "type": "function",
        "function": {
            "name": "get_current_stock_price",
            "description": "Get the current stock price for the given stock symbol",
            "parameters": {
                "type": "object",
                "properties": {
                    "symbol": {
                        "type": "string",
                        "description": "The stock symbol, e.g. AAPL, GOOGL, TSLA",
                    },
                    "exchange": {
                        "type": "string",
                        "description": "The stock exchange (optional)",
                        "enum": ["NYSE", "NASDAQ", "LSE", "TSX"],
                    },
                },
                "required": ["symbol"],
            },
        },
    },
]

response = client.chat.completions.create(
    model="moonshotai/Kimi-K2.5",
    reasoning={"enabled": False},
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": "What is the stock price of the company from the image",
                },
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://53.fs1.hubspotusercontent-na1.net/hubfs/53/image8-2.jpg",
                    },
                },
            ],
        },
    ],
    tools=tools,
)

print(
    json.dumps(
        response.choices[0].message.model_dump()["tool_calls"], indent=2
    )
)
The model analyzes the image to identify the company, then returns a function call with the appropriate stock symbol:
JSON
[
  {
    "id": "call_85951e7547ec4b81954b35e5",
    "type": "function",
    "function": {
      "name": "get_current_stock_price",
      "arguments": "{\"symbol\": \"GOOGL\"}"
    },
    "index": -1
  }
]