CyberBench v1.1

Can autonomous agents craft PoC inputs that trigger OSS-Fuzz vulnerabilities—and produce source patches that fix them?

ModelsModels with a result.48
Top resultBest result on this benchmark.80.4%Muse Wonder
Top-3 spreadPoints from first to third.2.5 pts
UpdatedDate the source changed the results.5 Nov 2026

Result and cost

48 of 48 results

Results

48 results
#ModelResultThe score from the source.CostUS dollars to run one task.TimeTime to run one task.
01Muse Wonder80.4% ±5.1$3.6614 min
02GPT-6 Sol78.0% ±5.1$1.3412 min
03Gemini 4 Argon77.9% ±5.3$5.1119 min
04GPT-5.6 Sol76.3% ±5.2$3.0511 min
04GPT-6 Luna76.3% ±5.3$0.13416 min
06Claude Haiku 5.575.5% ±5.2$1.2533 min
07MiMo-V2.6-Flash75.4% ±5.4$0.04628 min
08Kimi K375.2% ±5.6$2.1331 min
09Claude Opus 5.5 Range74.6% ±5.3$4.1723 min
10DeepSeek V4.1 Flash 101773.7% ±5.5$0.10514 min
11Muse Wonder Full73.6% ±5.6$3.6335 min
11GPT-5.6 Luna73.6% ±5.6$0.40416 min
13GPT Daybreak Blue73.0% ±5.3$3.9712 min
14MiMo-V2.6-Pro72.9% ±5.5$0.09428 min
14DeepSeek V4 Pro 081372.9% ±5.5$0.59323 min
16Inkling 2 Small72.7% ±5.7$1.2724 min
16Muse Spark 1.3 Max72.7% ±5.7$3.3416 min
18GPT-5.6 Terra72.1% ±5.4$3.3117 min
18GLM-5.372.1% ±5.4$2.6930 min
18DeepSeek V4 Flash 073172.1% ±5.4$0.19315 min
21Grok 4.571.0% ±5.8$2.9514 min
22Claude Fable 5.170.4% ±5.4$4.0119 min
23Claude Sonnet 5 Range70.3% ±5.7$4.8423 min
24Muse Spark 1.269.5% ±5.6$2.1520 min
24Ember-169.5% ±5.6$2.4322 min
24Grok 4.769.5% ±5.7$8.5416 min
27Muse Spark 1.369.4% ±5.8$4.2625 min
28GLM-5.3-Flash67.9% ±5.6$0.28436 min
29Qwen3.8-27B67.8% ±5.7$0.9942 min
30Grok 4.667.7% ±5.9$4.0737 min
31Inkling66.8% ±6.0$1.4614 min
32Claude Opus 565.4% ±5.5$3.5319 min
32Hy4 preview65.4% ±5.5$0.45837 min
34MiniMax M365.0% ±6.1$3.4141 min
35Inkling-Small64.1% ±6.2$0.25220 min
36Claude Sonnet 561.9% ±5.7$1.8917 min
37Claude Sonnet 5.559.6% ±5.2$2.9134 min
38Gemini 3.5 Flash-Lite59.1% ±6.2$0.42518 min
39Ling 3.0 Flash Af Rc358.3% ±6.1$0.10419 min
40Claude Opus 5.555.4% ±5.1$2.8518 min
41Mercury 2.545.5% ±6.1$0.12519 min
42Gemini 3.6 Flash45.4% ±3.7$1.3111 min
43Gemini 3.7 Flash43.8% ±2.2$2.158 min
43Gemini 3.8 Flash43.8% ±2.2$1.9610 min
45GPT-6 Astra41.1% ±2.6$1.844 min
46GPT-6.1 Sol39.3% ±2.7$0.2854 min
47Nemotron Lightning 3p5 30B A3B29.5% ±6.0$0.09351 min
48Qwen3.8 Max28.6% ±3.3$0.78510 min