Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Chat on Arena-Hard Style-Controlled
Loading...
46.1
Win-rate
PROSPER
41.836
42.943
44.05
45.157
Feb 22, 2026
Win-rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Win-rate
PROSPER
Backbone=Qwen2.5-7B-In...
2026.02
46.1
PROSPER-VB
Backbone=Qwen2.5-7B-In...
2026.02
45.5
RLCF
Backbone=Qwen2.5-7B-In...
2026.02
45
Qwen2.5-7B-Instruct
Model Scale=7B, LLM Ju...
2026.02
44.2
PROSPER-JC
Backbone=Qwen2.5-7B-In...
2026.02
42
Feedback
Search any
task
Search any
task