OpenAI Launches GPT-5.5 and GPT-5.5 Pro in the API

OpenAI has added GPT-5.5 to the Chat Completions API and Responses API, and also released GPT-5.5 Pro for harder problems that require more compute, together in the Responses API.

GPT-5.5 supports the following features.

• 1 million token context window
• Image input
• Structured outputs and function calling
• Prompt caching, Batch, tool search
• Built-in computer use, hosted shell, apply patch
• Skills, MCP, web search

The default reasoning effort value is medium, and when image_detail is unset or auto, existing behavior is preserved. Caching only works with extended prompt caching, and in-memory prompt caching is not supported.

Also, as of the April 21 change, GPT Image 2 was released as the latest model for image generation and editing. This model offers flexible image sizes, high-fidelity image input, token-based pricing, and a 50% discount on the Batch API.