← Models

GPT-5-Codex

OpenAI · GPT-5

Earlier GPT-5 Codex release entry, superseded by newer GPT-5.3-Codex in the current lineup.

Part of GPT family · Other versions: GPT-5.3-Codex , GPT-5.3 , GPT-5.4-Pro , GPT-5.4 , GPT-5.2-Codex , GPT-5.2-Pro , GPT-5.2 , GPT-5 nano , GPT-5 mini , GPT-5 , GPT-4.1 , GPT-4o mini Transcribe , GPT-4o mini TTS , GPT-4o Transcribe , GPT-4o mini , GPT-4o
Type
language
Context
400K tokens
Max Output
128K tokens
Status
legacy
Input
$1.25/1M tok
Output
$10/1M tok
API Access
Yes
License
proprietary
coding agentic software-engineering tool-use code-review refactoring
Released September 2025 · Updated March 6, 2026

Overview

Freshness note: Model capabilities, limits, and pricing can change quickly. This profile is a point-in-time snapshot last verified on February 26, 2026.

GPT-5-Codex was the initial coding-specialized release in the GPT-5 generation. It provided the foundation for OpenAI’s code-first routing strategy before newer GPT-5.x Codex variants became the recommended current path.

Capabilities

It is designed for multi-file code understanding, patch generation, refactoring, and review-oriented development workflows where reliability matters.

Technical Details

This generation entry exposed 400K context and 128K maximum output with support for tool-use and structured orchestration APIs.

Pricing & Access

Historical public API pricing for this release entry was:

  • Input: $1.25 per 1M tokens
  • Output: $10.00 per 1M tokens

For active production routing, use the latest Codex variant documented by OpenAI.

Best Use Cases

Treat this entry as a historical baseline for GPT-5 Codex evolution. For new coding-agent builds, prefer GPT-5.3-Codex or the broader GPT-5.4 tier when you need stronger mixed-task reasoning.

Comparisons

  • GPT-5.3-Codex (OpenAI): Current named successor in this snapshot.
  • GPT-5.4 (OpenAI): Broader high-end option when coding specialization is not required.
  • Claude Opus 4.6 (Anthropic): Strong alternative for complex coding and review tasks.