Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code Generation on LiveCodeBench v6 (mean@8, pass@8)
Loading...
61.82
Mean@8
MOPD
27.4792
36.3946
45.31
54.2254
May 12, 2026
Mean@8
Pass@8
Updated 21d ago
Evaluation Results
Method
Method
Links
Mean@8
Pass@8
MOPD
Model Backbone=Qwen3-8B
2026.05
61.82
67.23
MOPD
Model Backbone=Qwen3-4B
2026.05
57.01
65.48
SDPO
Model Backbone=Qwen3-8B
2026.05
49.49
64.34
SDPO
Model Backbone=Qwen3-4B
2026.05
48.84
63.23
GRPO
Model Backbone=Qwen3-8B
2026.05
43.65
58.72
GRPO
Model Backbone=Qwen3-4B
2026.05
40.75
55.43
Base
Model Backbone=Qwen3-8B
2026.05
30.97
53.02
Base
Model Backbone=Qwen3-4B
2026.05
28.8
49.36
Feedback
Search any
task
Search any
task