Instruction Following
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95ISR
9
Feb 26, 2026
59.7Win Rate vs ChatGPT
9
Feb 26, 2026
71.1Win Rate (vs ChatGPT)
9
Feb 26, 2026
76.8Win Rate (vs ChatGPT)
9
Feb 26, 2026
54.4Win Rate (vs ChatGPT)
9
Feb 26, 2026
65.8Win Rate (vs ChatGPT)
9
Feb 26, 2026
68.8Win Rate (vs ChatGPT)
9
Feb 26, 2026
57Win Rate (vs ChatGPT)
9
Feb 26, 2026
53.7Win Rate (vs ChatGPT)
9
Feb 26, 2026
53Win Rate (vs ChatGPT)
9
Feb 26, 2026
50.3Win Rate (vs ChatGPT)
9
Feb 26, 2026
47Win Rate (vs ChatGPT)
9
Feb 26, 2026
51.96Average Score
8
Jul 7, 2026
74.19DRFR Score
8
Jul 7, 2026
19.6Accuracy (Prompt)
8
Jun 25, 2026
12.224Reward Score
8
Jun 9, 2026
4.38Constraint Score
8
Jun 4, 2026
82.81Loose Accuracy (Prompt-level)
8
Jun 23, 2026
8.32Latency (ms/tok)
8
Jun 2, 2026
57.3Pr. (S)
8
May 29, 2026
1.3AH2 Score
8
May 19, 2026
37.98WB-Score
8
May 19, 2026
85.2Accuracy
8
May 19, 2026
72.66Accuracy
8
May 13, 2026
59.52IF-pS
8
May 12, 2026