GPT-5.3-Codex
OpenAI · released Feb 24, 2026
GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results...
Specification
- Context window
- 400K
- Max output
- 128K
- Knowledge cutoff
- Not stated
- Parameters
- Undisclosed
- Licence
- Proprietary
- Serving providers
- 2
- Moderated
- Yes
- Uptime
- 100.0%
Intelligence
32.5
84th percentile
Coding
—
Coding Index
Agentic
—
Agentic Index
Output speed
55 t/s
Median across providers
Latency
2.48s
Time to first token
Cost per task
$1.21
Estimated
Benchmarks
Where the score comes from
The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.
Evaluation scores
Percentage correct · higher is better
- GPQA Diamond91.5%
- τ²-bench (Telecom)86.0%
- AA-LCR (long context)83.3%
- IFBench75.4%
- Terminal-Bench Hard53.0%
- Humanity's Last Exam42.5%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| GPQA Diamond | 91.5% |
| τ²-bench (Telecom) | 86.0% |
| AA-LCR (long context) | 83.3% |
| IFBench | 75.4% |
| Terminal-Bench Hard | 53.0% |
| Humanity's Last Exam | 42.5% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- KAT-Coder-Pro V2KwaiPilot33.9
- Qwen3.8 27BQwen33.7
- Muse Spark 1.1Meta33.7
- GLM 5.2Z.ai33.7
- Gemini 3.5 FlashGoogle32.6
- GPT-5.3-CodexOpenAI32.5
- 32.5
- Ring-2.6-1TInclusionAI31.1
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| KAT-Coder-Pro V2 | 33.9 |
| Qwen3.8 27B | 33.7 |
| Muse Spark 1.1 | 33.7 |
| GLM 5.2 | 33.7 |
| Gemini 3.5 Flash | 32.6 |
| GPT-5.3-Codex | 32.5 |
| Qwen3 Max Thinking | 32.5 |
| Ring-2.6-1T | 31.1 |
Percentile among all indexed models
Pricing
What it costs to run
List prices per million tokens, plus what one representative task works out to.
List price
- Input / 1M tokens
- $1.75
- Output / 1M tokens
- $14
- Cached input / 1M
- $0.175
- Blended 3:1
- $4.81
One task, estimated
$1.21
- Input tokens
- 50,000
- Output tokens
- 80,000
- Profile
- Reasoning
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).
Arena
Head-to-head generation quality
Elo from pairwise judgements, broken out by the kind of thing the model was asked to build.
- Overall Elo
- 1,164
- Win rate
- 48.6%
- Strongest at
- Websites
- Tournaments
- 8,722
Elo by category
Dot position on a 975–1,175 scale · Elo has no meaningful zero
- 1,164
- 1,164
- 1,164
- 1,139
- 1,127
- 1,008
- 1,051
- 983
Agent categories (full-stack apps, mobile apps) are only scored for models tested in agent harnesses.
View as table
| Category | Elo | Win rate |
|---|---|---|
| Websites | 1164 | 48.6% |
| Data visualisation | 1164 | 50.6% |
| Game development | 1164 | 51.3% |
| UI components | 1139 | 47.3% |
| SVG | 1127 | 54.0% |
| 3D scenes | 1008 | 35.3% |
| Mobile apps | 1051 | 41.4% |
| Full-stack apps | 983 | 36.5% |