Vals CyberOffense v0.3.0 - Overall: leaderboard

Vals CyberOffense measures how many objectives agents complete in two controlled cybersecurity environments. The overall score averages the two environment completion rates. Mean range completion (%).

Metric: Mean range completion (%). Source: www.vals.ai. Saturation forecast: Around January 2027. 7 models tracked.

Top models

#ModelScoreOverall rank
1GPT-6 Sol80.85#18
2DeepSeek V4.1 Flash (1017)72.76#28
3Kimi K3.1 Pro63.21#16
4Kimi K337.56#35

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=vals-cyberoffense-v0-3-0-overall · How It Works · Data refreshed daily, snapshot 2026-10-29.