List your company
ResearchFormerly WizWand

Reward Modeling on LitBench

80.7AccuracyAC-GenRM
46.3855.2964.273.11Jan 12, 2026Jan 26, 2026Feb 9, 2026Feb 23, 2026Mar 9, 2026Mar 23, 2026Apr 7, 2026
Updated 6mo ago

Evaluation results

MethodLinks
AC-GenRMBackbone=Qwen3-8B2026.0480.7
AC-GenRMBackbone=Qwen3-4B2026.0479.6
AC-GenRMBackbone=Qwen3-1.7B2026.0477.6
claude-3-7-sonnet-20250219Model Category=Base Models (LLM-as-a-Judge)2026.0176.31
gpt-5-2025-08-07Model Category=Base Models (LLM-as-a-Judge)2026.0176.17
RM-NLHF-Qwen-32BModel Category=Our Generative Reward Models, Backbone=Qwen-32B2026.0174.92
Claude-Sonnet-3.7Category=Baselines2026.0473.1
AC-GenRMBackbone=Qwen3-0.6B2026.0472.8
o3-2025-04-16Model Category=Base Models (LLM-as-a-Judge)2026.0172.02
gemini-2.5-proModel Category=Base Models (LLM-as-a-Judge)2026.0171.81
ArmoRM-Llama3-8B-v0.1Model Category=Scalar Reward Models, Backbone=Llama3-8B2026.0171.57
INF-ORM-Llama3.1-70BModel Category=Scalar Reward Models, Backbone=Llama-3.1-70B2026.0171.41
DeepSeek-R1Category=Baselines2026.0471
GPT-4.1Category=Baselines2026.0470.2
deepseek-r1-0528Model Category=Base Models (LLM-as-a-Judge)2026.0170
DeepSeek-V3Category=Baselines2026.0470
RM-R1-Qwen-32BModel Category=Specialized Generative Reward Models, Backbone=Qwen-32B2026.0169.34
BaseBackbone=Qwen3-8B2026.0468.8
qwen3-maxModel Category=Base Models (LLM-as-a-Judge)2026.0168.59
qwen-plus-latestModel Category=Base Models (LLM-as-a-Judge)2026.0168.35
Claude-3.5-HaikuCategory=Baselines2026.0467.5
RRM-Qwen-32BModel Category=Specialized Generative Reward Models, Backbone=Qwen-32B2026.0167.46
BaseBackbone=Qwen3-4B2026.0466.6
Skywork-Reward-Llama-3.1-8B-v0.2Model Category=Scalar Reward Models, Backbone=Llama-3.1-8B2026.0165.93
RM-NLHF-Qwen-7BModel Category=Our Generative Reward Models, Backbone=Qwen-7B2026.0165.83
deepseek-v3Model Category=Base Models (LLM-as-a-Judge)2026.0165.36
gpt-4o-latestModel Category=Base Models (LLM-as-a-Judge)2026.0163.19
URM-LLaMa-3.1-8BModel Category=Scalar Reward Models, Backbone=Llama-3.1-8B2026.0163.02
GPT-4.1-MiniCategory=Baselines2026.0463
R1-Distill-Qwen-32BModel Category=Base Models (LLM-as-a-Judge), Backbone=Qwen-32B2026.0162.06
R1-Distill-Llama-8BModel Category=Base Models (LLM-as-a-Judge), Backbone=Llama-8B2026.0158.22
RM-R1-Qwen-7BModel Category=Specialized Generative Reward Models, Backbone=Qwen-7B2026.0157.5
BaseBackbone=Qwen3-1.7B2026.0454.3
DeepSeek-R1-Distill-Qwen-7BModel Category=Base Models (LLM-as-a-Judge), Backbone=Qwen-7B2026.0154.18
RRM-Qwen-7BModel Category=Specialized Generative Reward Models, Backbone=Qwen-7B2026.0153.83
BaseBackbone=Qwen3-0.6B2026.0447.7