OpenAI: GPT-5.2 Codex API
GPT-5.2-Codex is OpenAI's most advanced model optimized for long-horizon, agentic coding tasks. It integrates the professional knowledge of the GPT-5.2 family with specialized capabilities for software engineering, such as repo-aware intelligence and native context compaction. This model features an "agentic execution loop" that allows it to generate, test, and refine code autonomously in sandboxed environments. With its "xhigh" reasoning effort setting, it can work independently for hours on complex refactors, migrations, and cybersecurity audits, maintaining high steerability and temporal consistency across thousands of files.
- Context window: 400,000 tokens
- Max output: 128,000 tokens
- Input: text, image
- Output: text
- Reasoning: Supported
- Tool calling: Supported
- File input: Supported
- Released: 2025-12-11
- Knowledge cutoff: 2025-08-31
Frequently Asked Questions
What is the context window of GPT-5.2 Codex?
GPT-5.2 Codex supports a context window of up to 400,000 tokens.
Does GPT-5.2 Codex support function calling?
Yes. GPT-5.2 Codex supports tool / function calling.
Does GPT-5.2 Codex support reasoning?
Yes. GPT-5.2 Codex is a reasoning-capable model.
What is the knowledge cutoff of GPT-5.2 Codex?
The knowledge cutoff of GPT-5.2 Codex is 2025-08-31.