Codex Models by OpenAI
OpenAIProprietary
5 models2021–2026Up to 400k ctxFrom $1.75/1M input
Details
ResearcherOpenAI
LicenseProprietary
Commercial useCommercial use: conditional
Models5
Released2021–2026
Max context400k
Capabilities
Vision1 of 5 models
Reasoning1 of 5 models
Function Calling1 of 5 models
Tool Use1 of 5 models
Structured Outputs1 of 5 models
Code Execution1 of 5 models
Links
WebsiteAbout
OpenAI coding models optimized for agentic software engineering in Codex and similar developer environments, including GPT-5.3-Codex.
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view4 retired
GPT-5.3-CodexCurrent
Use when the workload needs code, 400k context, and reasoning.
2026-02code400k contextreasoning
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| GPT-5.3-Codex | Use when the workload needs code, 400k context, and reasoning. | 2026-02 | code400k contextreasoning | Current |
Release Timeline
4 release groups2026-02
1 current
GPT-5.3-Codex
Currentcode400k contextreasoning
2021-11
2 retired
Code Cushman 001
Archivedcode
Code Cushman 002
Archivedcode
2021-08
1 retired
Code Davinci 002
Archivedcode
2021-07
1 retired
Code Davinci 001
Archivedcode
Specifications(5 models)
| Model | Released | Context | Vision | Reasoning | Fn Calling | Tool Use | Structured Outputs | Code Exec |
|---|---|---|---|---|---|---|---|---|
| GPT-5.3-Codex | 2026-02 | 400k | Yes | Yes | Yes | Yes | Yes | Yes |
Available From(3 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| GPT-5.3-Codex | OpenRouter | $1.75 | $14 | Serverless |
| GPT-5.3-Codex | OpenAI API | $1.75 | $14 | Serverless |
| GPT-5.3-Codex | Vercel AI Gateway | $1.75 | $14 | Serverless |
Comparisons
All comparisons →Frequently Asked Questions
- What is Codex used for?
- Codex is used for code, vision and multimodal work, and reasoning. The family description and listed model capabilities point to those workloads as the best fit.
- How does Codex compare to GPT Realtime 2?
- Codex by OpenAI is strongest where you need code, while GPT Realtime 2 by OpenAI is the closest related family to check for realtime voice. Codex has 5 listed variants and reaches up to 400k context, while GPT Realtime 2 reaches up to 131k context, so compare the specs and pricing tables before choosing a production model.
- Which Codex model should I use?
- For the lowest listed input price, start with GPT-5.3-Codex through OpenAI API at $1.75/1M input tokens. For the most capable/latest local choice, evaluate GPT-5.3-Codex with 400k context and reasoning, tool use, function calling, structured outputs, and multimodal inputs.
