GPT-5.5
OpenAI · Released 24 Apr 2026 · gpt-5.5
$5.00
Input price / 1M tokens
#52 of 61, cheapest first
$30.00
Output price / 1M tokens
#57 of 61, cheapest first
1.05M
Context window
#2 of 71
86
Output tokens / second
#38 of 61
82.7%
Terminal-Bench 2.0
#1 of 7
58.6%
SWE-bench Pro
#10 of 16
78.7%
OSWorld-Verified
#4 of 8
84.4%
BrowseComp
#4 of 7
Summary
GPT-5.5 is a proprietary model from OpenAI, released on 24 Apr 2026.
At $5.00 input and $30.00 output per million tokens, it is #52 of 61 on input price, cheapest first.
Its 1.05M-token context window ranks #2 of 71.
Measured output speed is 86 tokens per second, #38 of 61.
A frontier model for coding and complex professional work, the flagship before GPT-5.6.
Benchmarks · 9 reported
| Benchmark | Score | Bar | Rank | Source |
|---|---|---|---|---|
| Terminal-Bench 2.0xhigh effort | 82.7% | #1 of 7 | OpenAI launch postSelf-reported | |
| SWE-bench Proxhigh effort; OpenAI notes labs have seen memorization on this eval | 58.6% | #10 of 16 | OpenAI launch postSelf-reported | |
| OSWorld-Verifiedxhigh effort | 78.7% | #4 of 8 | OpenAI launch postSelf-reported | |
| GDPvalwins or ties; xhigh effort | 84.9% | – | OpenAI launch postSelf-reported | |
| BrowseCompxhigh effort | 84.4% | #4 of 7 | OpenAI launch postSelf-reported | |
| GPQA Diamondxhigh effort | 93.6% | #4 of 27 | OpenAI launch postSelf-reported | |
| Humanity’s Last Exam (no tools)no tools; xhigh effort | 41.4% | #8 of 23 | OpenAI launch postSelf-reported | |
| Humanity’s Last Exam (with tools)with tools; xhigh effort | 52.2% | #16 of 16 | OpenAI launch postSelf-reported | |
| ARC-AGI-2xhigh effort | 85% | #2 of 6 | OpenAI launch postSelf-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-5.5, and the closest scores.
Terminal-Bench 2.0
#1 of 7Terminal tasks, 2025 set
- GPT-5.582.7
- GPT-5.475.1
- Gemini 3.1 Pro68.5
- Kimi K2.666.7
SWE-bench Pro
#10 of 16Harder multi-file engineering tasks
- GPT-5.558.6
- GPT-5.457.7
- Gemini 3.1 Pro54.2
- Kimi K2.658.6
- GLM-5.262.1
Compare GPT-5.5 side by side
Pre-filled with the two closest models. Swap any of them.
Sources
- price, tiers, long-context rule: developers.openai.com/api/docs/pricing, checked 2 Oct 2026
- context, max output, knowledge cutoff (Dec 1 2025), snapshot, modalities: developers.openai.com/api/docs/models/gpt-5.5, checked 2 Oct 2026
- API release date Apr 24 2026 (announced Apr 23), benchmarks, Fast/Priority 2.5x: openai.com/index/introducing-gpt-5-5/, checked 2 Oct 2026
- output speed: artificialanalysis.ai/models/gpt-5-5, checked 2 Oct 2026
Building on GPT-5.5?
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.