Remote Code Execution Attack Success Rate on OpenClaw
65C-F ScoreGPT-5.4 mini
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-5.4 miniFoundational LLM=GPT-5.4 mini2026.05 | 65 | 30 | 39.67 | |
| GLM-5Foundational LLM=GLM-52026.05 | 56.67 | 35.67 | 52.3 | |
| MiniMax-M2.7Foundational LLM=MiniMax-M2.72026.05 | 28 | 52.67 | 67.33 |