Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code reasoning on LCB v5
Loading...
33.72
Accuracy
MCPO-DAPO
13.6584
18.8667
24.075
29.2833
May 25, 2026
Accuracy
Updated 7d ago
Evaluation Results
Method
Method
Links
Accuracy
MCPO-DAPO
Backbone=Qwen3-8B
2026.05
33.72
MCPO-GRPO
Backbone=Qwen3-8B
2026.05
32.54
MT-GRPO
Backbone=Qwen3-8B
2026.05
30.33
MGS
Backbone=Qwen3-8B
2026.05
29.08
MCPO-DAPO
Backbone=Qwen3-4B
2026.05
28.93
MCPO-GRPO
Backbone=Qwen3-4B
2026.05
28.46
CLIPO
Backbone=Qwen3-8B
2026.05
28.46
DAPO
Backbone=Qwen3-8B
2026.05
28.04
GRPO
Backbone=Qwen3-8B
2026.05
27.73
MT-GRPO
Backbone=Qwen3-4B
2026.05
27.41
MGS
Backbone=Qwen3-4B
2026.05
27.04
DAPO
Backbone=Qwen3-4B
2026.05
25.88
GRPO
Backbone=Qwen3-4B
2026.05
25.79
CLIPO
Backbone=Qwen3-4B
2026.05
24.03
Base Model
Backbone=Qwen3-8B
2026.05
17.29
Base Model
Backbone=Qwen3-4B
2026.05
14.43
Feedback
Search any
task
Search any
task