Skip to content

Benchmark

ExploitGym

Coding benchmarks compare models on programming tasks. Scores are percentages.

Models measured
11
Scores recorded
11
Sources
1
IQ weight
1.87

Score distribution

26.2%
Mean
20.7%
Median
15.1%
Std dev
12.4%
10th percentile
48.5%
90th percentile
6%
Min
6%53.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.87 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
1Gemini 4 Carbon· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim53.5%
2GPT-6-cyber· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim48.5%
3GPT-6 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim42.4%
4GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim33.7%
5GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim23.2%
6Kimi K3.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim20.7%
7MiMo V2.6 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim17.8%
8DeepSeek V4.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim15.3%
9GLM 5.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim15%
10GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim12.4%
11MiMo V2.6 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim6%

Data credits: Data by LLM Stats