Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Chat on ArenaHard v2.0
Loading...
52
Win Rate
GOLF
3.7128
16.2489
28.785
41.3211
Mar 4, 2026
Win Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Win Rate
GOLF
Model Backbone=Qwen-3-8B
2026.03
52
Pairwise-GRPO
Model Backbone=Qwen-3-8B
2026.03
50.1
Rubric-as-Reward
Model Backbone=Qwen-3-8B
2026.03
50.08
Critique-GRPO
Model Backbone=Qwen-3-8B
2026.03
49.63
Direct-Likert
Model Backbone=Qwen-3-8B
2026.03
41.7
Qwen-3-8B
Model Backbone=Qwen-3-8B
2026.03
33.9
GOLF
Model Backbone=Llama-3...
2026.03
25.03
Rubric-as-Reward
Model Backbone=Llama-3...
2026.03
15.57
Critique-GRPO
Model Backbone=Llama-3...
2026.03
13.73
Pairwise-GRPO
Model Backbone=Llama-3...
2026.03
13.3
Direct-Likert
Model Backbone=Llama-3...
2026.03
11.73
Llama-3.1-8B-Instruct
Model Backbone=Llama-3...
2026.03
5.57
Feedback
Search any
task
Search any
task