Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Reasoning and Knowledge on Aggregate (BBH, MMLU, ARC-C, ThmQA)
Loading...
26.27
Mean Score
HybKD
14.154
17.2995
20.445
23.5905
May 25, 2026
Mean Score
Updated 7d ago
Evaluation Results
Method
Method
Links
Mean Score
HybKD
Distillation Pair=Llam...
2026.05
26.27
α-β divergence
Distillation Pair=Llam...
2026.05
25.01
JS divergence
Distillation Pair=Llam...
2026.05
24.84
Adaptive KL
Distillation Pair=Llam...
2026.05
24.41
Skew RKL
Distillation Pair=Llam...
2026.05
23.69
Skew FKL
Distillation Pair=Llam...
2026.05
23.37
Reverse KL
Distillation Pair=Llam...
2026.05
22.75
Total Variation
Distillation Pair=Llam...
2026.05
20.66
HybKD
Distillation Pair=Gemm...
2026.05
16.26
Skew FKL
Distillation Pair=Gemm...
2026.05
15.81
α-β divergence
Distillation Pair=Gemm...
2026.05
15.79
Reverse KL
Distillation Pair=Gemm...
2026.05
15.78
JS divergence
Distillation Pair=Gemm...
2026.05
15.75
Skew RKL
Distillation Pair=Gemm...
2026.05
15.08
Adaptive KL
Distillation Pair=Gemm...
2026.05
14.63
Total Variation
Distillation Pair=Gemm...
2026.05
14.62
Feedback
Search any
task
Search any
task