Benchmark
ExploitGym on-target exploits
Higher scores are better.
Benchmark sourceScores from ExploitGym leaderboard data docs.
- Models measured
- 16
- Scores recorded
- 27
- Sources
- 1
- IQ weight
- 1.24
Score distribution
- 144.1
- Mean
- 130
- Median
- 104.7
- Std dev
- 12
- 10th percentile
- 293
- 90th percentile
- 4
- Min
This benchmark has a weight of 1.24 for chat models, measured over 16 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | |
|---|---|---|
| 1 | GPT-6 Astraagent: unstated, eval: token-budgeted, version: unstated, focus: model, reported by: OpenAI, date: 2026-09-03· ExploitGym leaderboard data docs | 368 |
| 2 | GPT-6.1 Solagent: unstated, eval: token-budgeted, version: unstated, focus: model, reported by: OpenAI, date: 2026-09-29· ExploitGym leaderboard data docs | 305 |
| 3 | Claude Opus 5.5agent: unstated, eval: 6h timeout, version: unstated, focus: model, reported by: Anthropic, date: 2026-09-22· ExploitGym leaderboard data docs | 300 |
| 4 | GPT-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 docs | 293 |
| 5 | Claude Opus 5.5agent: unstated, eval: 2h timeout, version: unstated, focus: model, reported by: Anthropic, date: 2026-09-22· ExploitGym leaderboard data docs | 289 |
| 6 | DeepSeek 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 docs | 226 |
| 7 | GPT-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 docs | 216 |
| 8 | DeepSeek 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 docs | 206 |
| 9 | Claude Opus 5agent: unstated, eval: 6h timeout, version: unstated, focus: model, reported by: Anthropic, date: 2026-07-24· ExploitGym leaderboard data docs | 191 |
| 10 | DeepSeek 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 docs | 186 |
| 11 | Claude Opus 5agent: unstated, eval: 2h timeout, version: unstated, focus: model, reported by: Anthropic, date: 2026-07-24· ExploitGym leaderboard data docs | 171 |
| 12 | GLM 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 docs | 159 |
| 13 | GLM 5.3agent: DoGNAVY, eval: 10h timeout, selected subset, version: v1, focus: agent, reported by: deepsec@DARKNAVY, date: 2026-09-10· ExploitGym leaderboard data docs | 157 |
| 14 | GLM 5.3agent: Claude Code, eval: rescaled 6h timeout, version: unstated, focus: model, reported by: Z.ai, date: 2026-08-18· ExploitGym leaderboard data docs | 130 |
| 15 | GPT 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 docs | 129 |
| 16 | Claude Opus 4.8agent: unstated, eval: 6h timeout, version: unstated, focus: model, reported by: Anthropic, date: 2026-05-28· ExploitGym leaderboard data docs | 120 |
| 17 | GLM 5.3agent: Claude Code, eval: rescaled 2h timeout, version: unstated, focus: model, reported by: Z.ai, date: 2026-08-18· ExploitGym leaderboard data docs | 105 |
| 18 | Claude Opus 4.8agent: unstated, eval: 2h timeout, version: unstated, focus: model, reported by: Anthropic, date: 2026-05-28· ExploitGym leaderboard data docs | 80 |
| 19 | GLM 5.2agent: DoGNAVY, eval: selected subset, version: v1, focus: agent, reported by: deepsec@DARKNAVY, date: 2026-08-22· ExploitGym leaderboard data docs | 79 |
| 20 | GPT 5.4agent: Codex CLI, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs | 61 |
| 21 | GLM 5.2agent: Claude Code, eval: rescaled 6h timeout, version: unstated, focus: model, reported by: Z.ai, date: 2026-08-18· ExploitGym leaderboard data docs | 39 |
| 22 | GLM 5.2agent: Claude Code, eval: rescaled 2h timeout, version: unstated, focus: model, reported by: Z.ai, date: 2026-08-18· ExploitGym leaderboard data docs | 29 |
| 23 | Claude 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 docs | 16 |
| 24 | Claude Opus 4.7agent: Claude Code, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs | 12 |
| 25 | Gemini 3.1 Pro Previewagent: Gemini CLI, eval: 2h timeout, version: v0, focus: model, reported by: ExploitGym Team, date: 2026-05-13· ExploitGym leaderboard data docs | 12 |
| 26 | Muse 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 docs | 7 |
| 27 | GLM 5.1agent: Claude Code, eval: 2h timeout, version: v1, focus: model, reported by: ExploitGym Team, date: 2026-06-16· ExploitGym leaderboard data docs | 4 |
Tokens the model generated per task, as measured by the source. Comparable within this benchmark only.