GPT-5.5 Codex
OpenAI coding-specialized GPT-5.5 Codex model for long-horizon agentic coding tasks, repository edits, and test generation.
GPT-5.5 Codex is OpenAI’s coding-specialized model, purpose-built for long-horizon agentic coding tasks including repository-wide edits, test generation, multi-step debugging, and code review. Unlike the general-purpose GPT-5.5, Codex is optimized for maintaining coherence across long coding sessions and understanding the full context of a codebase.
With a 400,000-token context window, GPT-5.5 Codex can process entire repositories in a single pass, making it suitable for tasks that require understanding file relationships, import chains, and project-wide patterns. It uses the OpenAI Responses API and integrates with Codex-specific workflows for repository indexing and context management.
For teams building coding agents, GPT-5.5 Codex offers the deepest code understanding in the OpenAI lineup. It competes with specialized coding models like Devstral 2 from Mistral AI and Claude Opus 4.7 from Anthropic, with the advantage of native integration with OpenAI’s broader ecosystem including the Responses API, function calling, and structured output capabilities for agent orchestration.
| Provider | OpenAI |
|---|---|
| Context Window | 400000 |
| Pricing | Pricing and access can vary; verify against official OpenAI API docs before production use. |
| API Style | OpenAI Responses API |
| SDK | OpenAI SDK, Codex workflows |
| MCP | Works through Codex-like harnesses and tool adapters. |
| Agent | Strong fit for long-running coding agents, repository edits, test iteration, and code review. |
| RAG | Useful for codebase RAG when paired with repository indexing and source-grounded context. |
| Source Freshness | recently_verified |
| Version Status | current |
| Version Boundary | GPT-5.5 Codex is the latest OpenAI coding-specialized model, updated alongside the GPT-5.5 family (April 2026). It inherits the 400K-token context window and improved reasoning of GPT-5.5 while maintaining coding-specialized optimizations. |
Key Facts
- GPT-5.5 Codex is the latest coding-specialized model in the OpenAI lineup.
- Optimized for long-running coding agents, repository-wide edits, and test generation.
- 400K-token context window enables full-repository awareness.
- Use the Responses API and Codex workflows for integration.
Best For
Not Ideal For
Capability Matrix
| Capability | Status |
|---|---|
| Coding | Very Strong |
| Agentic Coding | Very Strong |
| Tool Use | Strong |
| Reasoning | Strong |
SEO
| SEO Title | GPT-5.5 Codex API, Pricing, SDK, MCP & Agent Compatibility |
|---|---|
| Description | GPT-5.5 Codex by OpenAI: OpenAI coding-specialized GPT-5.5 Codex model for long-horizon agentic coding tasks, repository edits, and test generation. |
| Canonical | /model/gpt-5-2-codex |
| Updated | 2026-05-21 |
Compare
| Comparison | Compared With |
|---|---|
| GPT-5.5 Codex vs Claude Sonnet 4.6 | Claude Sonnet 4.6 |
| GPT-5.5 Codex vs Devstral 2 | Devstral 2 |
| GPT-5.5 Codex vs Claude Opus 4.7 | Claude Opus 4.7 |
Compatibility Facts
| Layer | Target | Status | Evidence | Updated |
|---|---|---|---|---|
| api | openai-responses-api | verify_required | GPT-5.2 Codex includes OpenAI model documentation and OpenAI SDK GitHub sources. | 2026-05-18 |
FAQ
| What is GPT-5.5 Codex? | GPT-5.5 Codex is the current OpenAI coding-optimized model for long-horizon agentic coding tasks with 400K-token context. Confirm current access, pricing, and limits in official OpenAI documentation. |
|---|---|
| What is GPT-5.5 Codex best for? | GPT-5.5 Codex is best for coding, agent workflow, agentic coding, code review. |
| How should GPT-5.5 Codex be verified before production use? | Check current pricing, availability, limits, and API behavior against the listed official and GitHub sources. This entry was updated on 2026-05-21. |
| How are model version boundaries represented? | Model version boundaries are represented by stable slugs, versionStatus, and versionNote so detail pages and feeds can distinguish current, preview, legacy, or unknown entries. |
| How should agentic coding models be compared? | Agentic coding models should be compared by repository-context handling, tool execution reliability, code review quality, SDK support, and whether the provider publishes stable model identifiers. |
| Which model fields matter most for code review selection? | For code review, compare coding capability, reasoning capability, source freshness, SDK support, relationship facts, and related code-review scenario pages. |
Relationship Facts
| Source | Type | Target | Confidence |
|---|---|---|---|
| gpt-5-2-codex | best_for | agentic-coding | 0.84 |
| gpt-5-2-codex | best_for | code-review | 0.8 |
| gpt-5-2-codex | supports | openai-responses-api | 0.74 |
Sources
| Name | Type | Citation | Last Verified |
|---|---|---|---|
| OpenAI API Models | docs | Official OpenAI model documentation is the primary source for Codex model availability and model-family positioning. | 2026-05-21 |
| OpenAI Node SDK | github | GitHub SDK reference for OpenAI JavaScript and TypeScript API integration. | 2026-05-21 |
External Resources
- OpenAI API Models — Official OpenAI model documentation is the primary source for Codex model availability and model-family positioning.
- OpenAI Node SDK — GitHub SDK reference for OpenAI JavaScript and TypeScript API integration.