← Models

GPT-5.3-Codex

OpenAI · GPT-5

Latest Codex-tuned GPT-5 model for repository-scale implementation, review, and agent workflows.

Part of GPT family · Other versions: GPT-5.3 , GPT-5.4-Pro , GPT-5.4 , GPT-5.2-Codex , GPT-5.2-Pro , GPT-5.2 , GPT-5-Codex , 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
current
Input
$1.5/1M tok
Output
$6/1M tok
API Access
Yes
License
proprietary
coding agentic software-engineering tool-use code-review refactoring
Released March 2026 · 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 March 6, 2026.

GPT-5.3-Codex is OpenAI’s current Codex-tuned model in the GPT-5 line. OpenAI positions it as the newer coding-specific successor to earlier GPT-5.x Codex variants, with stronger repo-scale implementation and patch-generation performance.

Capabilities

The model is built for multi-file edits, code review, test-aware implementation loops, and longer agent workflows where reliable tool use matters. It is especially suitable when prompts define acceptance criteria, repository constraints, and validation steps up front.

Technical Details

OpenAI’s model docs list GPT-5.3-Codex with a 400K context window and 128K max output. OpenAI also describes GPT-5.4 as building on GPT-5.3-Codex for coding and agentic tasks, which makes GPT-5.3-Codex a strong current reference point for software engineering workflows.

Pricing & Access

Published API pricing (per 1M tokens):

  • Input: $1.50
  • Output: $6.00

OpenAI lists GPT-5.3-Codex in the API and also references availability through Codex surfaces.

Best Use Cases

Use GPT-5.3-Codex for repository-level implementation, migration work, CI-adjacent code review automation, and coding agents that need stronger patch reliability than general-purpose model routes.

Comparisons

  • GPT-5.4 (OpenAI): Better general high-end model for mixed professional workflows, but not as explicitly coding-positioned.
  • GPT-5.4-Pro (OpenAI): Premium depth tier for highest-stakes work at much higher cost.
  • Claude Code + Claude Opus 4.6 (Anthropic): Strong alternative coding stack when Anthropic-native tooling is preferred.