GPT-5.2 Codex is OpenAI's dedicated coding model built on the GPT-5.2 architecture. Optimized for software engineering tasks with strong agentic coding capabilities and expanded context window support.
- Dedicated coding model built on GPT-5.2 architecture for software engineering tasks.
- 400,000-token context window with 128,000 max output tokens and reasoning token support.
- Optimized for agentic coding workflows including code generation, debugging, and refactoring.
- Supports function calling, structured outputs, and streaming for developer integration.
- Pricing is $1.75 per 1M input tokens and $14 per 1M output tokens (cached input at $0.175).
Model Information
Supported Formats
GPT 5.5OpenAI's most advanced model released April 2026. GPT-5.5 is recommended for most reasoning workloads, combining frontier intelligence with configurable reasoning effort (none → xhigh). Available in API, Codex, and ChatGPT.
GPT 5.5 ProHighest-intelligence API option in the GPT-5.5 family. Use for the most challenging problems that require maximum reasoning depth and can tolerate higher latency. Supports full reasoning effort range from none to xhigh.
GPT 5.4GPT-5.4 is OpenAI's frontier model for complex professional work, designed for agentic, coding, and professional workflows with the highest reasoning capability. Features a 1.05M token context window, native computer use, and built-in support for web search, file search, and MCP.