Compare AI models
| Benchmarks | ||
CyberGymCybersecurity — Tests the AI on cybersecurity challenges — finding and exploiting software weaknesses inside a safe sandbox. Higher is better. | 77.5% | 85.6% |
| Overview | ||
| Company | OpenAI | |
| Release date | Jul 21 2026 | Jun 22 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Gemini 3.5 Flash CybervsClaude Haiku 5.5GPT-5.5-CybervsClaude Haiku 5.5Gemini 3.5 Flash CybervsMuse Spark OpenGPT-5.5-CybervsMuse Spark OpenGemini 3.5 Flash CybervsGrok 4.7GPT-5.5-CybervsGrok 4.7Gemini 3.5 Flash CybervsDeepSeek-V4.1-Flash-1017GPT-5.5-CybervsDeepSeek-V4.1-Flash-1017Gemini 3.5 Flash CybervsMistral Large 4GPT-5.5-CybervsMistral Large 4Gemini 3.5 Flash CybervsKimi K3GPT-5.5-CybervsKimi K3Frequently asked questions
GPT-5.5-Cyber leads Gemini 3.5 Flash Cyber on 1 of the 1 benchmark they both report (CyberGym). GPT-5.5-Cyber shipped 29 days before Gemini 3.5 Flash Cyber, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On CyberGym, GPT-5.5-Cyber leads at 85.6% vs Gemini 3.5 Flash Cyber at 77.5%.