GPT-5
OpenAI · released Aug 7, 2025
GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy...
Specification
- Context window
- 400K
- Max output
- 128K
- Knowledge cutoff
- Sep 30, 2024
- Parameters
- Undisclosed
- Licence
- Proprietary
- Serving providers
- 2
- Moderated
- Yes
- Uptime
- 100.0%
Intelligence
23.0
72th percentile
Coding
37.8
Coding Index
Agentic
—
Agentic Index
Output speed
58 t/s
Median across providers
Latency
4.23s
Time to first token
Cost per task
$0.86
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
- AIME 202594.3%
- MMLU-Pro87.1%
- GPQA Diamond85.4%
- τ²-bench (Telecom)84.8%
- LiveCodeBench84.6%
- AA-LCR (long context)78.2%
- IFBench73.1%
- Terminal-Bench Hard32.6%
- Humanity's Last Exam28.5%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| AIME 2025 | 94.3% |
| MMLU-Pro | 87.1% |
| GPQA Diamond | 85.4% |
| τ²-bench (Telecom) | 84.8% |
| LiveCodeBench | 84.6% |
| AA-LCR (long context) | 78.2% |
| IFBench | 73.1% |
| Terminal-Bench Hard | 32.6% |
| Humanity's Last Exam | 28.5% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- Claude Opus 4.5Anthropic23.7
- GLM 5V TurboZ.ai23.5
- Kimi K2.5Moonshot AI23.5
- GPT-5OpenAI23.0
- Qwen3.5-27BQwen22.9
- M2.7MiniMax22.8
- M2.5MiniMax22.8
- Hy3 previewTencent22.7
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| Claude Opus 4.5 | 23.7 |
| GLM 5V Turbo | 23.5 |
| Kimi K2.5 | 23.5 |
| GPT-5 | 23.0 |
| Qwen3.5-27B | 22.9 |
| M2.7 | 22.8 |
| M2.5 | 22.8 |
| Hy3 preview | 22.7 |
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
- $1.25
- Output / 1M tokens
- $10
- Cached input / 1M
- $0.125
- Blended 3:1
- $3.44
One task, estimated
$0.86
- 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
- 1,222
- Win rate
- 60.5%
- Strongest at
- Data visualisation
- Tournaments
- 233
Elo by category
Dot position on a 1,050–1,225 scale · Elo has no meaningful zero
- 1,222
- 1,188
- 1,187
- 1,186
- 1,181
- 1,056
Agent categories (full-stack apps, mobile apps) are only scored for models tested in agent harnesses.
View as table
| Category | Elo | Win rate |
|---|---|---|
| Data visualisation | 1222 | 60.5% |
| Game development | 1188 | 59.3% |
| SVG | 1187 | 64.1% |
| Websites | 1186 | 53.7% |
| UI components | 1181 | 57.4% |
| 3D scenes | 1056 | 41.3% |