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

Reward Modeling on RM-Bench Chat Hard

83.3Accuracy

Gemini-2.5-Flash

Updated 4d ago

Evaluation Results

Method	Links
Gemini-2.5-Flash 2026.02		83.3
RM-R1-32B (DeepSeek-Dist) 2026.02		83.1
RRM-32B 2026.02		81.1
Rubric-ARM-voting@5 2026.02		80.7
RM-R1-32B (Qwen-2.5-Inst) 2026.02		80.3
RM-R1-14B (Qwen-2.5-Inst) 2026.02		79.8
Rubric-ARM 2026.02		79.6
API (Rubric+Judge) 2026.02		79.2
RM-R1-14B (DeepSeek-Dist) 2026.02		78.9
Rubric-RM-voting@5 2026.02		75.4
Rubric-RM 2026.02		74.1
Claude-3.5-Sonnet 2026.02		74
API (direct Judge) 2026.02		71.2
RM-R1-7B (Qwen-2.5-Inst) 2026.02		70
RRM-7B 2026.02		69.5
RM-R1-7B (DeepSeek-Dist) 2026.02		67.3
Qwen-3-8B (Rubric+Judge) 2026.02		63.6
REWARDAGENT_MINI 2025.02		60.2
REWARDAGENT_MINI 2025.02		59.7
DeepSeek-R1-Distill-Llama-8B 2025.02		56.8
JudgeLRM-7B 2026.02		56.1
REWARDAGENT_LLAMA 2025.02		53.5
DeepSeek-R1 2025.02		50.1
REWARDAGENT_LLAMA 2025.02		49.9
o3-mini 2025.02		38.6
Skywork-Reward-Gemma-2-27B 2025.02		35.1
ArmoRM-Llama3-8B-v0.1 2025.02		34.6
Skywork-Reward-Llama-3.1-8B-v0.2 2025.02		31.8
GPT-4o 2025.02		27.9
internlm2-20b-reward 2025.02		26.1
INF-ORM-Llama3.1-70B 2025.02		25.1
Llama3-8B Instruct 2025.02		20.2
internlm2-7b-reward 2025.02		19.9
GPT-4o mini 2025.02		15