Remote Code Execution Attack Success Rate on Claude Code
43.67C-F Success RateGPT-5.4 mini
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-5.4 miniFoundational LLM=GPT-5.4 mini2026.05 | 43.67 | 11 | 5.33 | |
| MiniMax-M2.7Foundational LLM=MiniMax-M2.72026.05 | 42 | 44.33 | 64.33 | |
| GLM-5Foundational LLM=GLM-52026.05 | 37.67 | 17.33 | 2.67 |