Our new X account is live! Follow @wizwand_team for updates

Reward Modeling on Reward Bench V2 (Accuracy)

83.44Accuracy

gpt-5-2025-08-07

Updated 4d ago

Evaluation Results

Method	Links
gpt-5-2025-08-07 2026.01		83.44
INF-ORM-Llama3.1-70B 2026.01		80.66
o3-2025-04-16 2026.01		80.09
URM-LLaMa-3.1-8B 2026.01		80.04
Skywork-Reward-Llama-3.1-8B-v0.2 2026.01		79.07
RM-NLHF-Qwen-32B 2026.01		78.67
qwen-plus-latest 2026.01		77.17
claude-3-7-sonnet-20250219 2026.01		77.14
RRM-Qwen-32B 2026.01		73.4
deepseek-r1-0528 2026.01		72.92
RM-R1-Qwen-32B 2026.01		72.6
qwen3-max 2026.01		71.69
gemini-2.5-pro 2026.01		71.17
ArmoRM-Llama3-8B-v0.1 2026.01		68.97
gpt-4o-latest 2026.01		64.78
R1-Distill-Qwen-32B 2026.01		61.54
RM-NLHF-Qwen-7B 2026.01		57.57
deepseek-v3 2026.01		54.36
RRM-Qwen-7B 2026.01		50.82
RM-R1-Qwen-7B 2026.01		46.79
R1-Distill-Llama-8B 2026.01		40.95
DeepSeek-R1-Distill-Qwen-7B 2026.01		30.52