Benchmarks · safety
CyberBench v1.1
Agent performance reproducing OSS-Fuzz vulnerabilities with proof-of-concept inputs and patching the underlying bugs.
39 models · 40 results · updated 2026-11-05 · vals.ai ↗
Leaderboard
GPT-6 Sol max
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 27 more models under All models in settings.
More from this benchmark
1 published measuresTime per task48 results · lower is better
26 more models under All models in settings.
All results
40 rows| # | Model | Effort | CyberBench v1.1 | Subscription $ / task | API $ / task | Time |
|---|---|---|---|---|---|---|
| 1 | GPT-6 Sol | max | 78.0% | $0.041 | $1.34 | 12 min |
| 2 | Gemini 4 Argon | high | 77.9% | $3.64 | $5.11 | 19 min |
| 3 | GPT-5.6 Sol | max | 76.3% | $0.084 | $3.05 | 11 min |
| 4 | GPT-6 Luna | max | 76.3% | $0.010 | $0.13 | 16 min |
| 5 | Haiku 5.5 | max | 75.5% | $0.026 | $1.25 | 33 min |
| 6 | MiMo V2.6 Flash | high | 75.4% | free | $0.046 | 28 min |
| 7 | Kimi K3 | max | 75.2% | $0.19 | $2.13 | 31 min |
| 8 | DeepSeek V4.1 Flash | high | 73.7% | $0.008 | $0.11 | 14 min |
| 9 | GPT-5.6 Luna | max | 73.6% | $0.019 | $0.40 | 16 min |
| 10 | DeepSeek V4 Pro 0813 | max | 72.9% | $0.59 | $0.59 | 23 min |
| 11 | MiMo V2.6 Pro | max | 72.9% | $0.079 | $0.094 | 28 min |
| 12 | Inkling 2 Small | max | 72.7% | $1.27 | $1.27 | 24 min |
| 13 | Muse Spark 1.3 | max | 72.7% | $1.67 | $3.34 | 16 min |
| 14 | DeepSeek V4 Flash 0731 | high | 72.1% | $0.19 | $0.19 | 15 min |
| 15 | GLM 5.3 | max | 72.1% | $0.28 | $2.69 | 30 min |
| 16 | GPT-5.6 Terra | max | 72.1% | $0.091 | $3.31 | 17 min |
| 17 | Grok 4.5 | high | 71.0% | $0.46 | $2.95 | 14 min |
| 18 | Fable 5.1 | max | 70.4% | $0.63 | $4.01 | 19 min |
| 19 | Muse Spark 1.2 | xhigh | 69.5% | $1.07 | $2.15 | 20 min |
| 20 | Grok 4.7 | xhigh | 69.5% | $0.79 | $8.54 | 16 min |
| 21 | Muse Spark 1.3 | xhigh | 69.4% | $2.13 | $4.26 | 25 min |
| 22 | GLM 5.3 Flash | max | 67.9% | free | $0.28 | 36 min |
| 23 | Qwen 3.8 27B | xhigh | 67.8% | $0.072 | $0.99 | 42 min |
| 24 | Grok 4.6 | high | 67.7% | $0.42 | $4.07 | 37 min |
| 25 | Inkling | max | 66.8% | $0.13 | $1.46 | 14 min |
| 26 | Opus 5 | max | 65.4% | $0.061 | $3.53 | 19 min |
| 27 | Hy4 preview | high | 65.4% | $0.35 | $0.46 | 37 min |
| 28 | MiniMax M3 | max | 65.0% | $0.57 | $3.41 | 41 min |
| 29 | Inkling Small | max | 64.1% | $0.20 | $0.25 | 20 min |
| 30 | Sonnet 5 | max | 61.9% | $0.033 | $1.89 | 17 min |