Benchmark · KnowledgeCompetitive

Exploitbench

Updated 2026-04-24
Models tested
8
Top score
47.4
GPT-5.5
Median
21.0
min 13.3
Top-5 spread
σ 9.9
wide open

Best score over time · one chart, every benchmark

EXPLOITBENCH8 MODELS · FRONTIER RUNNING MAX0255075100SCORE ↑Oct 25Dec 25Jan 26Mar 26Apr 26RELEASE DATE →benchgecko.ai/benchmark/exploitbench · frontier
Frontier on Exploitbench rose from 13.7 to 47.4 in 6 months · +33.7 points · latest leader GPT-5.5 from OpenAI.
Pink dots = frontier records · 5 totalClick to open model page

8 models tested · sorted by score

What it measures · how it is scored · sources · as of 2026-11-17

  • How it is scored · Scored as a percentage · higher is better · maximum 100
  • Released · 2026-05-13
  • Scores from · Epoch AI Benchmarking Hub (8 scores)

Same category · related evaluations