Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Reward Modeling on Multimodal Reward Bench
Loading...
85.62
Reward Bench Score
Proxy-GRM-RL
40.2552
52.0326
63.81
75.5874
Mar 17, 2026
Reward Bench Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Reward Bench Score
Proxy-GRM-RL
Proxy Agent=Proxy-SFT,...
2026.03
85.62
Proxy-GRM-RL
Proxy Agent=None, Data...
2026.03
85.28
Proxy-GRM-SFT
Proxy Agent=None, Data...
2026.03
85.18
Proxy-GRM-RL
Proxy Agent=Proxy-RL,...
2026.03
84.78
R1-Reward
Proxy Agent=None, Data...
2026.03
82.2
Claude-3.7-Sonnet
Proxy Agent=None, Data...
2026.03
71.9
Claude-3.5-Sonnet-(2024-06-22)
Proxy Agent=None, Data...
2026.03
71.5
Qwen2-VL-72B
Proxy Agent=None, Data...
2026.03
70.9
GPT-4o-(2024-08-06)
Proxy Agent=None, Data...
2026.03
70.8
IXC-2.5-Reward
Proxy Agent=None, Data...
2026.03
66.6
VITA-1.5
Proxy Agent=None, Data...
2026.03
53.6
SliME
Proxy Agent=None, Data...
2026.03
42
Feedback
Search any
task
Search any
task