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

Reward Modeling on PPE-Preference

79.8Accuracy

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

Updated 1mo ago

Evaluation Results

Method
Skywork-Reward-V2-Llama-3.1-8B-40M 2025.07	79.8	-	-
Skywork-Reward-V2-Llama-3.1-8B 2025.07	77.3	-	-
Skywork-Reward-V2-Qwen3-8B 2025.07	70.6	-	-
Gemini-2.5-Pro 2026.02	69.8	8.3	64
Skywork-Reward-V2-Qwen3-4B 2025.07	69.5	-	-
Claude-Sonnet-4.5 2026.02	69.2	14.5	59.1
Skywork-Reward-V2-Llama-3.2-3B 2025.07	69.1	-	-
Claude-Sonnet-4.5-thinking 2026.02	68.7	8.7	62.7
GPT-4o 2025.07	67.7	-	-
Skywork-Reward-V2-Qwen3-1.7B 2025.07	67.6	-	-
Claude-3.5-Sonnet 2025.07	67.3	-	-
DeepSeek-GRM-27B 2025.07	67.2	-	-
J1-Llama-70B 2025.07	67	-	-
DeepSeek-V3.2-thinking 2026.02	66.8	26.2	49.2
GPT-5-chat 2026.02	66.6	19.6	53.5
Skywork-Reward-V2-Llama-3.2-1B 2025.07	66.6	-	-
DeepSeek-V3.2-chat 2026.02	66.4	24	50.5
J1-Llama-70B 2025.07	66.3	-	-
GenRM-RLVR-14B 2026.02	65.7	45.7	35.6
GenRM-R-Align-14B 2026.02	65.7	26.9	48.1
RM-R1-Qwen-32B 2026.02	65.6	30	45.9
DeepSeek-GRM-27B 2025.07	65.3	-	-
Skywork-Reward-V2-Qwen3-0.6B 2025.07	65.3	-	-
Qwen3-14B 2026.02	65.2	36.9	41.1
RRM-32B 2026.02	65.1	62.4	24.5
RM-R1-DS-32B 2026.02	64.9	46.5	34.7
GPT-5-thinking 2026.02	64.7	6.9	58.8
GPT-OSS-120B 2026.02	64.5	40.7	38.2
Llama-3.1-Nemotron-70B 2025.07	64.2	-	-
INF-ORM-Llama3.1-70B 2025.07	64.2	-	-
GenRM-RLVR-8B 2026.02	63.7	47.4	33.5
GenRM-R-Align-8B 2026.02	63.6	34.9	41.4
Skywork-Reward-Gemma-2-27B-v0.2 2025.07	63.6	-	-
GRM-llama3-8B-distill 2025.07	63.2	-	-
FsfairX-LLaMA3-RM-v0.1 2025.07	63.1	-	-
Starling-RM-34B 2025.07	62.8	-	-
Gemini-3-Pro 2026.02	62.7	1.6	61.6
LDL-Reward-Gemma-2-27B-v0.1 2025.07	62.4	-	-
Qwen3-8B 2026.02	62.2	45.2	34.1
Skywork-Reward-Llama-3.1-8B-v0.2 2025.07	62.2	-	-
Internlm2-7b-reward 2025.07	62.1	-	-
GRM-Llama3-8B-rewardmodel-ft 2025.07	62.1	-	-
Skywork-Reward-Llama-3.1-8B 2025.07	62.1	-	-
RM-Mistral-7B 2025.07	61.8	-	-
BTRM Qwen2 7b 0613 2025.07	61.8	-	-
Qwen3-4B-Thinking-2507 2026.02	61.5	33.7	40.8
Internlm2-20b-reward 2025.07	61	-	-
ArmoRM-Llama3-8B-v0.1 2025.07	60.6	-	-
QRM-Llama3.1-8B 2025.07	60.6	-	-
J1-Llama-8B 2025.07	60.6	-	-
Qwen3-4B-Instruct-2507 2026.02	60.4	44.8	33.3
Skywork-Reward-Gemma-2-27B 2025.07	60.3	-	-
J1-Llama-8B 2025.07	60.3	-	-
URM-LLaMa-3.1-8B 2025.07	60.2	-	-
GRM-gemma2-2B-rewardmodel-ft 2025.07	59.7	-	-
Eurus-RM-7b 2025.07	59.6	-	-
Llama-3-OffsetBias-RM-8B 2025.07	59.2	-	-
internlm2-1.8b-reward 2025.07	57.3	-	-
QRM-Llama3.1-8B-v2 2025.07	57.2	-	-
QRM-Gemma-2-27B 2025.07	52.3	-	-