GPT-5
OpenAI · Released 7 Aug 2025 · gpt-5
$1.25
Input price / 1M tokens
#27 of 61, cheapest first
$10.00
Output price / 1M tokens
#39 of 61, cheapest first
400K
Context window
#52 of 71
75
Output tokens / second
#44 of 61
74.9%
SWE-bench Verified
#6 of 12
96.7%
τ²-bench
#1 of 6
Summary
GPT-5 is a proprietary model from OpenAI, released on 7 Aug 2025.
At $1.25 input and $10.00 output per million tokens, it is #27 of 61 on input price, cheapest first.
Its 400K-token context window ranks #52 of 71.
Measured output speed is 75 tokens per second, #44 of 61.
The previous-generation reasoning model for coding and agentic tasks, launched August 2025. OpenAI now points users to newer models.
Benchmarks · 3 reported
| Benchmark | Score | Bar | Rank | Source |
|---|---|---|---|---|
| SWE-bench Verifiedn=477 (23 of 500 tasks omitted); short verification-focused prompt | 74.9% | #6 of 12 | OpenAI launch post for developersSelf-reported | |
| Aider polyglotdiff format | 88% | – | OpenAI launch post for developersSelf-reported | |
| τ²-benchhigh reasoning effort | 96.7% | #1 of 6 | OpenAI launch post for developersSelf-reported |
Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.
Similar models
Models with the most benchmarks in common with GPT-5, and the closest scores.
SWE-bench Verified
#6 of 12Real GitHub issues, fixed and tested
- GPT-574.9
- Gemini 3 Flash78
- Gemini 3.1 Pro80.6
- Claude Haiku 4.573.3
- Claude Sonnet 4.577.2
- Mistral Medium 3.577.6
Compare GPT-5 side by side
Pre-filled with the two closest models. Swap any of them.
Sources
- price, tiers: developers.openai.com/api/docs/pricing, checked 2 Oct 2026
- context, max output, knowledge cutoff (Sep 30 2024), snapshot: developers.openai.com/api/docs/models/gpt-5, checked 2 Oct 2026
- shutdown date Dec 11 2026, replacement: developers.openai.com/api/docs/deprecations, checked 2 Oct 2026
- release date, 272K input / 128K output limits, benchmarks: openai.com/index/introducing-gpt-5-for-developers/, checked 2 Oct 2026
- output speed: artificialanalysis.ai/models/gpt-5, checked 2 Oct 2026
Building on GPT-5?
Get the architecture right before the bill arrives.
We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.