Skip to content

Benchmark

ExploitGym on-target exploits (userspace)

Higher scores are better.

Models measured
11
Scores recorded
15
Sources
1
IQ weight
0.90

Score distribution

63
Mean
70
Median
48.3
Std dev
4
10th percentile
119
90th percentile
4
Min
4149
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.90 for chat models, measured over 11 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-5.6 Solagent: Codex CLI, model note: reasoning max, eval: 6h timeout, version: v1, focus: model, reported by: OpenAI & ExploitGym Team, date: 2026-07-13· ExploitGym leaderboard data docs149
2GPT-5.6 Solagent: Codex CLI, model note: reasoning max, eval: 2h timeout cutoff, version: v1, focus: model, reported by: OpenAI & ExploitGym Team, date: 2026-07-13· ExploitGym leaderboard data docs119
3DeepSeek V4.1 Flashagent: VARAS-OneMind (智安·宇盾), eval: 6h timeout, version: v1, focus: agent, reported by: VARAS, date: 2026-09-24· ExploitGym leaderboard data docs115
4DeepSeek V4.1 Flashagent: MopMonk Agent, eval: 8h timeout, selected subset, version: v1, focus: agent, reported by: Mopmonk AI, date: 2026-11-04· ExploitGym leaderboard data docs107
5GLM 5.3agent: DoGNAVY, eval: 10h timeout, selected subset, version: v1, focus: agent, reported by: deepsec@DARKNAVY, date: 2026-09-10· ExploitGym leaderboard data docs96
6GLM 5.3agent: VARAS-OneMind (智安·宇盾), eval: 6h timeout, version: v1, focus: agent, reported by: VARAS, date: 2026-09-11· ExploitGym leaderboard data docs91
7DeepSeek V4.1 Flashagent: MopMonk Agent, eval: 8h timeout, selected subset, version: v1, focus: agent, reported by: Mopmonk AI, date: 2026-10-21· ExploitGym leaderboard data docs90
8GPT 5.5agent: Codex CLI, model note: OpenAI's default safety filters block all GPT-5.5 exploit attempts under default prompting., eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs70
9GPT 5.4agent: Codex CLI, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs38
10GLM 5.2agent: DoGNAVY, eval: selected subset, version: v1, focus: agent, reported by: deepsec@DARKNAVY, date: 2026-08-22· ExploitGym leaderboard data docs36
11Claude Opus 4.6agent: Claude Code, model note: Results obtained in collaboration with Anthropic., eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs12
12Gemini 3.1 Pro Previewagent: Gemini CLI, eval: 2h timeout, version: v0, focus: model, reported by: ExploitGym Team, date: 2026-05-13· ExploitGym leaderboard data docs10
13Claude Opus 4.7agent: Claude Code, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs4
14GLM 5.1agent: Claude Code, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs4
15Muse Spark 1.1agent: Meta Agent (off-the-shelf agent harness), model note: helpful-only versoin, eval: 4h timeout, version: v1, focus: model, reported by: Meta AI, date: 2026-07-09· ExploitGym leaderboard data docs4

Data credits: ExploitGym leaderboard (Wang et al., UC Berkeley), arXiv:2605.11086. No licence is stated for the result data; the repository's Apache-2.0 licence covers its code only, and task data keeps its upstream licences. Literal exploit counts; hidden rows omitted.