GPT-5.3-Codex

Released

GPT-5.3-Codex is an AI model released by OpenAI on Thursday, Feb 5 2026, 56 days after GPT-5.2. Benchmark results (shown below) cover BullshitBench v2, SWE-Bench Verified, Next.js Evals, and OSWorld-Verified.

API pricing

Input
$1.75
Cached input
$0.175
Output
$14.00
  • Reasoning or thinking is supported.
USD per 1M tokens

Available from

Azure$1.75$14.00400K
USD per 1M tokens

Benchmarks

Coding

SWE-Bench Verified
85%
#7 of 57
Next.js Evals
81%
#5 of 25

Agentic & tool use

OSWorld-Verified
64.7%
#23 of 28

Robustness

BullshitBench v2
24%
#50 of 74

Source: BenchLM, retrieved 31 August 2026. Every other score here is the figure the lab published at launch.

Compare GPT-5.3-Codex with

GPT-5.3-Codex

Suggested comparisons

Frequently asked questions

GPT-5.3-Codex was released by OpenAI on Thursday, Feb 5 2026.

GPT-5.3-Codex was built by OpenAI. Creators of ChatGPT and the GPT series of models. Pioneered large-scale language model research.

GPT-5.3-Codex costs $1.75 per million input tokens and $14.00 per million output tokens through the OpenAI API. Cached input is $0.175 per million tokens. Rates are pay-as-you-go API prices verified against OpenAI's published pricing on August 18, 2026.

GPT-5.3-Codex reports 4 tracked benchmark scores — BullshitBench v2: 24%; SWE-Bench Verified: 85%; Next.js Evals: 81%; OSWorld-Verified: 64.7%. Scores are the figures published at release by OpenAI.

No. GPT-5.3-Codex is a proprietary model. The weights are not published — it is available only through the provider's own API, apps, or partner platforms.

OpenAI's previous tracked release was GPT-5.2 on Dec 11 2025, 56 days earlier. It was followed by GPT-5.3-Codex-Spark on Feb 12 2026.

All OpenAI releases

43 tracked

2022

1 release