Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Overall Performance Evaluation on Aggregate All Domains
Loading...
38.14
Average Accuracy
VRPO
23.1848
27.0674
30.95
34.8326
Aug 5, 2025
Average Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Average Accuracy
VRPO
Model Scale=Qwen3-8B
2025.08
38.14
λ-GRPO
Model Scale=Qwen3-8B
2025.08
37.95
Reinforce++
Model Scale=Qwen3-8B
2025.08
36.6
Dr.GRPO
Model Scale=Qwen3-8B
2025.08
36.59
GRPO
Model Scale=Qwen3-8B
2025.08
35.19
Base
Model Scale=Qwen3-8B
2025.08
34.64
VRPO
Model Scale=Qwen3-1.7B
2025.08
34.56
PPO
Model Scale=Qwen3-8B
2025.08
34.55
KTAE
Model Scale=Qwen3-8B
2025.08
34.55
Reinforce++
Model Scale=Qwen3-1.7B
2025.08
33.26
λ-GRPO
Model Scale=Qwen3-1.7B
2025.08
32.68
PPO
Model Scale=Qwen3-1.7B
2025.08
32.51
GRPO
Model Scale=Qwen3-1.7B
2025.08
32.43
Dr.GRPO
Model Scale=Qwen3-1.7B
2025.08
32.19
Base
Model Scale=Qwen3-1.7B
2025.08
30.78
KTAE
Model Scale=Qwen3-1.7B
2025.08
23.76
Feedback
Search any
task
Search any
task