Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Reasoning on MMLU-Pro (zero-shot latest)
Loading...
69.65
Accuracy (zero-shot)
MIXED-CUTS
63.566
65.1455
66.725
68.3045
Apr 20, 2026
Accuracy (zero-shot)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (zero-shot)
MIXED-CUTS
Backbone=Qwen3-4B, Tra...
2026.04
69.65
Standard GRPO
Backbone=Qwen3-4B, Tra...
2026.04
68.59
Base Model
Backbone=Qwen3-4B, Tra...
2026.04
63.8
Feedback
Search any
task
Search any
task