Skip to content

Benchmark

ExploitGym on-target exploits (V8)

Higher scores are better.

Models measured
11
Scores recorded
15
Sources
1
IQ weight
0.78

Score distribution

33.6
Mean
38
Median
25.4
Std dev
2
10th percentile
70
90th percentile
0
Min
071
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.78 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 docs71
2DeepSeek 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 docs70
3GPT-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 docs59
4DeepSeek 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 docs58
5DeepSeek 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 docs57
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 docs46
7GLM 5.3agent: DoGNAVY, eval: 10h timeout, selected subset, version: v1, focus: agent, reported by: deepsec@DARKNAVY, date: 2026-09-10· ExploitGym leaderboard data docs44
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 docs38
9GLM 5.2agent: DoGNAVY, eval: selected subset, version: v1, focus: agent, reported by: deepsec@DARKNAVY, date: 2026-08-22· ExploitGym leaderboard data docs23
10GPT 5.4agent: Codex CLI, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs22
11Claude Opus 4.7agent: Claude Code, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs8
12Claude 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 docs3
13Muse 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 docs3
14Gemini 3.1 Pro Previewagent: Gemini CLI, eval: 2h timeout, version: v0, focus: model, reported by: ExploitGym Team, date: 2026-05-13· ExploitGym leaderboard data docs2
15GLM 5.1agent: Claude Code, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs0

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.