Skip to main content

ExploitGym

A controlled benchmark for evaluating whether AI agents can extend vulnerability-triggering inputs into working exploits.

Data verified

Benchmark score on ExploitGym — July 9, 2026

BenchLM mirrors the published score view for ExploitGym. Claude Mythos 5 leads the public snapshot at 17.5% , followed by GPT-5.5 (13.4%) and GPT-5.4 (6.0%). BenchLM does not use these results to rank models overall.

4 modelsAgenticCurrentDisplay onlyUpdated July 9, 2026

The published ExploitGym snapshot is tightly clustered at the top: Claude Mythos 5 sits at 17.5%, while the third row is only 11.5 points behind. The broader top-10 spread is 16.7 points, so the benchmark still separates strong models even when the leaders cluster.

4 models have been evaluated on ExploitGym. The benchmark falls in the Agentic category. This category carries a 22% weight in BenchLM.ai's overall scoring system. ExploitGym is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About ExploitGym

Year

2026

Tasks

898 exploitation tasks

Format

Working exploit generation

Difficulty

Advanced cybersecurity exploitation

ExploitGym contains 898 containerized tasks sourced from real-world vulnerabilities across userspace programs, V8, and the Linux kernel. BenchLM normalizes successful exploit counts to percentage of the 898-task suite and keeps the benchmark display-only because it measures dual-use offensive cybersecurity capability.

BenchLM freshness & provenance

Version

ExploitGym 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Benchmark score table (4 models)

1
Claude Mythos 5claude-mythos-5
17.5%
2
GPT-5.5gpt-5-5
13.4%
3
GPT-5.4gpt-5-4
6.0%
4
Muse Spark 1.1muse-spark-1-1
0.8%

FAQ

What does ExploitGym measure?

A controlled benchmark for evaluating whether AI agents can extend vulnerability-triggering inputs into working exploits.

Which model scores highest on ExploitGym?

Claude Mythos 5 by Anthropic currently leads with a score of 17.5% on ExploitGym.

How many models are evaluated on ExploitGym?

4 AI models have been evaluated on ExploitGym on BenchLM.

Last updated: July 9, 2026 · BenchLM version ExploitGym 2026

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.