Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-term state poisoning evaluation on OpenClaw ZH
Loading...
3.82
Harm Score (HS)
GPT-4o
3.7344
4.3122
4.89
5.4678
May 7, 2026
Harm Score (HS)
Updated 23d ago
Evaluation Results
Method
Method
Links
Harm Score (HS)
GPT-4o
2026.05
3.82
Grok-1
2026.05
4.94
Kimi k2.5
2026.05
4.98
MiniMax-abab6.5
2026.05
5.96
Feedback
Search any
task
Search any
task