Skip to main content
Use the Responses API for reasoning, tool calling, image and file inputs, structured output, and typed SSE streaming. Function and custom tools, computer actions, local shell, apply patch, and client tool search run in your application. Remote MCP and web search run upstream. Use caller-owned Files, vector stores, Code Interpreter, hosted shell containers, image generation, and background responses. Stored responses support retrieval, cancellation, stream resume, and previous_response_id. Chat Completions supports function calling only with reasoning_effort: none. Reasoning: none, low, medium (default), high, xhigh, max.

Model ID

Pass this exact public ID in the top-level model field.

Supported modes

  • text_generation

Inputs

  • Text
  • Image
  • File

Outputs

  • Text

Capabilities

  • Text generation
  • Reasoning
  • Tool calling

Additional capabilities

  • Function calling
  • Custom tools
  • Web search
  • Remote MCP
  • Computer use
  • Tool search
  • Local shell
  • Apply patch
  • Structured outputs
  • Encrypted reasoning replay
  • Code Interpreter
  • Hosted shell and containers
  • File search
  • Responses image generation
  • Background responses See Function calling for tool declarations and result submission.

Technical details

Availability and final customer prices can change. Check the Dashboard model catalog before production use.
See the OpenAI Chat guide for Chat Completions, Responses, tools, files and background conversations. Try GPT-6 Sol in API Explorer. See the API reference overview for authentication, request conventions, and response schemas.