Does tavily support python openai function calling for llm?

Other Tavily Topics
1

do you have to write a JSON schema yourself in tools and pass it to llm?
the tavily search function has so many parameters and it takes a lot of effort to write the schema.
as far as i know if you import tavily to langchain, you dont need to do this.

2

Hi there,

Thanks for reaching out.

You don’t have to manually write the entire JSON schema for every Tavily function. We’ve provided ready-to-use schemas for search, extract, map, and crawl so you can just copy-paste them into your workflow. This saves time and lets you expose only the parameters you care about.
Two easy ways to do this:

  • Copy from our docs – just grab the ready-made schema from the Tavily docs and drop it into your tools definition.(Note: The schemas we provide are designed for the OpenAI Responses API, but we’ve also added guidance in the docs on how to adapt them for the Chat Completions API.)

  • Use the OpenAI Agents SDK – skip schema definitions entirely by wrapping tavily.search() with @function_tool, and the SDK handles the JSON schema automatically (we’ve added an example for the Agents SDK in the docs as well)

Please refer to our docs: https://docs.tavily.com/documentation/integrations/openai

Best,
May