Docs
Search
Ctrl K
Models
Playground
Compare
Providers
Apps
Rankings
Models
Playground
Compare
Providers
Apps
Rankings
Docs
Search
Ctrl K
Sign In
Sign In
ExploitBench (Cap%) - Benchmark Leaderboard & Model Performance | AI Stats
ExploitBench (Cap%)
Overview
Overview
Type: percentage
Cybersecurity
Recorded Results
1
Average Score
78%
Score Range
78% - 78%
Leading Model
78% - Claude Fable 5
Scores Over Time
Individual benchmark scores plotted by date.
Models Using This Benchmark
Organisation
Model
Reported
Top Score
Info
Self Reported
Source
Anthropic
Claude Fable 5
09 Jun 2026
78%
Cap%; starred row in Anthropic's launch table
Yes
Source