Share your thoughts, 1 month free Claude Pro on usSee more

Reward Modeling on RMB

89.3Accuracy

Skywork-Reward-V2-Llama-3.1-8B-40M

Updated 8d ago

Evaluation Results

Method	Links
Skywork-Reward-V2-Llama-3.1-8B-40M 2025.07		89.3
Skywork-Reward-V2-Llama-3.1-8B 2025.07		86.4
DeepSeek-V3.2 2026.03		83.9
Skywork-Reward-V2 2025.07		83.9
FARE-8B 2026.03		83.2
Skywork-Reward-V2-Qwen3-8B 2025.07		81.2
Skywork-Reward-V2-Qwen3-4B 2025.07		80.6
Skywork-Reward-V2-Llama-3.2-3B 2025.07		80.5
Mix-GRM 2026.03		80.1
Gemini-3-Flash 2026.03		79.2
Base-GRM 2026.03		79.2
ConsistRM (Qwen3-8B) 2026.04		79.1
Mix-GRM 2026.03		78.9
Qwen3-8B 2026.04		78.8
RubricRM-8B 2026.03		78.5
Base-GRM 2026.03		78.5
Skywork-Reward-V2 2025.07		78.3
Qwen3-8B + RFT 2026.04		78.2
Skywork-Reward-V2-Qwen3-1.7B 2025.07		78.1
Qwen3-8B + CAAR 2026.04		78.1
Qwen3-4B + CAAR 2026.04		77.5
Skywork-Reward-V2 2025.07		77.1
Skywork-Reward-V2 2025.07		77.1
Qwen3-8B + SFT 2026.04		76.9
Skywork-Reward-V2-Llama-3.2-1B 2025.07		76.7
Skywork-Reward-V2 2025.07		76.7
Qwen3-4B + RFT 2026.04		76.7
Qwen3-4B + SFT 2026.04		76.5
ConsistRM (Qwen3-4B) 2026.04		76.5
Qwen3-4B 2026.04		76.4
OPRM-Qwen2.5-32B 2026.02		75.6
Qwen3-4B + TTRL 2026.04		75.6
OPRM-RgFT-Qwen2.5-32B 2026.02		74.8
Skywork-Reward-V2-Qwen3-0.6B 2025.07		74.5
Skywork-Reward-8B 2026.03		74.4
OPRM-RgFT-Qwen2.5-72B 2026.02		74.2
Qwen3-8B + TTRL 2026.04		74.2
GPT-4o 2026.02		73.8
OPRM-Qwen2.5-14B 2026.02		73.8
GPT-4o 2025.07		73.8
GPT-4o-0806 2025.05		73.8
OPRM-Qwen2.5-72B 2026.02		73.5
JudgeLRM-7B 2026.03		73.1
RM-R1-Qwen-Instruct-32B 2025.07		73
RM-R1-QWEN-INSTRUCT-32B 2025.05		73
OPRM-RgFT-Qwen2.5-14B 2026.02		72.8
Starling-RM-34B 2025.07		72
OPRM-Qwen2.5-7B 2026.02		71.5
DeepSeek-GRM-16B 2026.03		70.8
Claude-3.5-sonnet 2026.02		70.6
Claude-3.5-Sonnet 2025.07		70.6
Claude-3-5-sonnet-20240620 2025.05		70.6
INF-ORM-Llama3.1-70B 2025.07		70.5
INF-ORM-Llama3.1-70B 2025.07		70.5
INF-ORM-Llama3.1-70B 2025.05		70.5
DeepSeek-GRM-27B 2025.07		70.3
DeepSeek-GRM-27B 2026.04		70.3
FsfairX-LLaMA3-RM-v0.1 2025.07		70.2
GRM-Llama3-8B-rewardmodel-ft 2025.07		70.2
CIR-AMS/BTRM_Qwen2_7b_0613 2025.07		70.2
OPRM-RgFT-Qwen2.5-7B 2026.02		70.1
Nemotron-4-340B-Reward 2026.02		69.9
RM-R1-7B (Instruct) 2026.03		69.9
Nemotron-4-340B-Reward 2025.05		69.9
RM-R1-DeepSeek-Distill-Qwen-32B 2025.07		69.8
RM-R1-DEEPSEEK-DISTILLED-QWEN-32B 2025.05		69.8
Skywork-Reward-Gemma-2-27B 2025.07		69.5
Skywork-Reward-Gemma-2-27B-v0.2 2025.07		69.4
Skywork-Reward-Llama-3.1-8B 2025.05		69.3
Skywork-Reward-Llama-3.1-8B 2025.07		69.2
RM-R1-QWEN-INSTRUCT-14B 2025.05		69.2
DeepSeek-GRM-27B 2026.02		69
DeepSeek-GRM-27B 2025.07		69
DeepSeek-GRM-27B 2025.05		69
LLaMA-3.1-70b-Instruct 2026.02		68.9
Llama3.1-70B-Instruct 2025.05		68.9
GRM-llama3-8B-distill 2025.07		68.8
RM-R1-DEEPSEEK-DISTILLED-QWEN-14B 2025.05		68.5
Eurus-RM-7b 2025.05		68.3
GRM-gemma2-2B-rewardmodel-ft 2025.07		68
LDL-Reward-Gemma-2-27B-v0.1 2025.07		67.9
Internlm2-7b-reward 2025.07		67.1
Internlm2-7b-reward 2025.05		67.1
DeepSeek-GRM-27B-RFT 2026.02		67
DeepSeek-GRM-27B-RFT 2025.05		67
Self-taught-evaluator-llama3.1-70B 2025.05		67
RM-Mistral-7B 2025.07		66.6
Skywork-Reward-Llama-3.1-8B-v0.2 2025.07		66.6
RM-R1-QWEN-INSTRUCT-7B 2025.05		66.4
DeepSeek-V2.5-0905 2026.02		65.7
URM-LLaMa-3.1-8B 2025.07		65.7
Eurus-RM-7b 2025.07		65.5
Ray2333/GRM-gemma2-2B-rewardmodel-ft 2025.07		65.5
Skywork-Critic-Llama-3.1-70B 2025.05		65.5
RM-R1-7B (Distill) 2026.03		65.1
Llama-3.1-Nemotron-70B 2025.07		64.9
LLM-as-a-Judge 2026.02		64.8
QRM-Llama3.1-8B 2025.07		64.7
ArmoRM-8B-v0.1 2026.02		64.6
ArmoRM-Llama3-8B-v0.1 2025.07		64.6

Showing 100 of 120 rows