Skip to main content

Overview

OpenAI’s Responses API is their new recommended API for inference, offering improved performance and features compared to Chat Completions. Key features:
  • 3% better performance on reasoning tasks
  • 40-80% improved cache utilization
  • Built-in tools: web search, file search, code interpreter
  • Server-side conversation state via previous_response_id
  • Semantic streaming events
Inference feature compatibility:
  • tool calling (supported)
  • native JSON object response_format (supported)
  • native JSON schema response_format (recommended)
  • Instructor markdown-JSON fallback (fallback)

Example