LLM Reference

Codex Models by OpenAI

OpenAIProprietary
5 models2021–2026Up to 400k ctxFrom $1.75/1M input

Last refreshed 2026-06-10. Next refresh: weekly.

Details

ResearcherOpenAI
LicenseProprietary
Commercial useCommercial use: conditional
Models5
Released2021–2026
Max context400k

Capabilities

Vision1 of 5 models
Reasoning1 of 5 models
JSON / Tool use1 of 5 models
Structured Outputs1 of 5 models
Code Execution1 of 5 models

Links

Website

About

OpenAI coding models optimized for agentic software engineering in Codex and similar developer environments, including GPT-5.3-Codex.

Decision facts

Best fit
codevision and multimodal workreasoning
Capability starting point
GPT-5.3-Codex with 400k context and reasoning, JSON / Tool use, structured outputs, and multimodal inputs
Lowest tracked input
GPT-5.3-Codex · $1.75/1M · OpenRouter
Closest related family
GPT-5

Current Variants

Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.

1 in view4 retired

Use when the workload needs code, 400k context, and reasoning.

2026-02code400k contextreasoning

Release Timeline

4 release groups
2026-02
1 current
GPT-5.3-Codex
code400k contextreasoning
Current
2021-11
2 retired
Archived
Archived
2021-08
1 retired
Archived
2021-07
1 retired
Archived

Specifications(5 models)

Codex model specifications comparison
ModelReleasedContextVisionReasoningJSON / Tool useStructured OutputsCode Exec
GPT-5.3-Codex2026-02400kYesYesYesYesYes

Available From(3 providers)

Pricing

Codex model pricing by provider
ModelProviderInput / 1MOutput / 1MType
GPT-5.3-CodexOpenRouter$1.75$14Serverless
GPT-5.3-CodexOpenAI API$1.75$14Serverless
GPT-5.3-CodexVercel AI Gateway$1.75$14Serverless

Popular comparisons in this family

Models(5)