Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Instruction Following on Harness-MU Conflict n=368
Loading...
97.35
Accuracy
Deepseek
77.2884
82.4967
87.705
92.9133
Jun 20, 2026
Accuracy
W.Acc
Per-User Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
W.Acc
Per-User Accuracy
Deepseek
Framework=Harness-MU,...
2026.06
97.35
97.35
17.43
Qwen
Framework=Harness-MU,...
2026.06
95.81
95.81
17.88
Gemini
Framework=Harness-MU,...
2026.06
92.25
92.25
16.5
GPT
Framework=Harness-MU,...
2026.06
78.06
78.06
14.03
Feedback
Search any
task
Search any
task
