Ctrl K
Back to models

OpenAI

GPT-5.3-Codex

  • Released
  • Closed weights
Compare
Intelligence Index

Extra high setting

32.5
Estimated · Artificial Analysis v4.3.2
API price from
$1.75 in / $14.00 out / 1M tokOpenAI · direct API
Context window
400K tok

Performance profile

ReasoningRank #15 of 107
91.5%
GPQA Diamond

Capabilities

Context window

Modalities

Reasoning modes

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
X-HighEstimated
Intelligence Index
32.5

Pricing & access

API route
OpenAIAPI
Tier & contextStandard

USD per 1M tokens

Input
$1.75
Output
$14.00
Cached input
$0.175
Responses API tool compatibility (opens in a new tab)
  • Web search
  • Shell
  • Skills

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

GPT-5.3-Codex · Extra high#21 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 18Grok 4.7Extra high43.1%
  6. 19Qwen3.8 Max (0902)Reasoning43.1%
  7. 20GPT-5.6 TerraMax42.9%
  8. 21GPT-5.3-CodexExtra high42.5%
  9. 22Qwen3.8 2.4T A95BReasoning42.4%
  10. 23GLM-5.3Max42.3%
  11. 24DeepSeek-V4-Pro-0813Max41.0%

Peers use their best measured result on this benchmark.