GPT-5
OpenAI's August 2025 unified frontier model that auto-routes between a fast model and a deeper "thinking" variant.
- Publisher
- OpenAI
- Family
- GPT 5
- Params
- Undisclosed (closed)
- Context
- 400K
- Input
- Output
- Providers
- openai
- License
- proprietary
- Released
- 7 Aug 2025
- Canonical
- openai.com/index/introducing-gpt-5
Cite
Notes
Only stored in your browser.
Context
400K
Weights
Closed
API only
Input
$1.25
/1M tok
Output
$10
/1M tok
Cache read
$0.13
0.10x input
Price & speed via Artificial Analysis · ranked across tracked models · caching via OpenRouter
| Reasoning effort | Index | Price | Speed |
|---|---|---|---|
| high | 23.0 | $3.44 | 0 t/s |
| medium | 22.9 | $3.44 | 0 t/s |
| low | 20.8 | $3.44 | 0 t/s |
| minimal | 11.4 | $3.44 | 0 t/s |
| ChatGPT | 10.4 | $0.00 | 0 t/s |
Intelligence Index per reasoning-effort setting, via Artificial Analysis
Reported on 62 evals across 15 domains - best 100.0% on Gutenberg Env · #1 of 5
Reported eval scores
62Mastermind1.6
MCP Universe92.7
MCP Universe92.7
Fingpt Sentiment91.7
Github MCP88.9
Capa68.1
Vpct 154.8
Hypothesis Forge48.9
Head-to-heads
29benchmark comparisons against other models
vs GPT-4.1 Mini41 shared evalsvs GPT-5 Mini36 shared evalsvs Claude Sonnet 4.531 shared evalsvs GPT-5 Nano28 shared evalsvs GPT-4.128 shared evalsvs Claude 4 Sonnet25 shared evalsvs Gemini 2.5 Pro24 shared evalsvs GPT-5.223 shared evalsvs Claude Opus 4.523 shared evalsvs GPT-4.1 Nano21 shared evalsvs Grok 4.1 Fast20 shared evalsvs Gemini 3.1 Pro Preview19 shared evalsvs Claude Opus 4.619 shared evalsvs GPT-5.419 shared evalsvs DeepSeek V3.219 shared evalsvs GPT-4o19 shared evalsvs GPT-5.518 shared evalsvs Kimi K2.518 shared evalsvs Claude Opus 4.817 shared evalsvs Claude Opus 4.717 shared evalsvs gpt-oss-120b17 shared evalsvs GLM 4.717 shared evalsvs Claude Fable 516 shared evalsvs GPT-4o-mini16 shared evalsvs Claude Sonnet 4.616 shared evalsvs GLM 5.116 shared evalsvs GPT-5.4 Mini16 shared evalsvs Qwen3.6 Plus15 shared evalsvs GPT-6 Astra12 shared evals