Inference APIs

Can I use tool calling and JSON mode?

Yes, on GPT-OSS 120B, DeepSeek V4 / V4.1 Flash, GLM 5.3 Flash and Llama 3.3 70B. Pass tools in the OpenAI format and read tool_calls from the response; pass response_format: {"type": "json_object"} for JSON mode. Streamed tool calls arrive as delta.tool_calls fragments.

Related

More questions