Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multiple-choice Reasoning on MMLU
Loading...
32.8
Accuracy
FineWeb-Edu
29.784
30.567
31.35
32.133
Apr 17, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
FineWeb-Edu
2026.04
32.8
NAG_Qwen3-1.7B
Target Setting=Single-...
2026.04
32.2
NAG_Llama-3.2-3B
Target Setting=Single-...
2026.04
31.5
NAG_Qwen3-1.7B
Target Setting=Multi-T...
2026.04
31.5
NAG_Llama-3.2-3B
Target Setting=Multi-T...
2026.04
31.4
NAG_SmolLM3-3B
Target Setting=Single-...
2026.04
31.2
BETR
Target Setting=Single-...
2026.04
31.1
NAG_SmolLM3-3B
Target Setting=Multi-T...
2026.04
30.6
Random
2026.04
30.2
BETR
Target Setting=Multi-T...
2026.04
29.9
Feedback
Search any
task
Search any
task