Reward Modeling on Reward Bench V2 (Accuracy)
83.44Accuracygpt-5-2025-08-07
Evaluation results
| Method | Links | |
|---|---|---|
| gpt-5-2025-08-07Model Category=Base Models (LLM-as-a-Judge)2026.01 | 83.44 | |
| INF-ORM-Llama3.1-70BModel Category=Scalar Reward Models, Backbone=Llama-3.1-70B2026.01 | 80.66 | |
| o3-2025-04-16Model Category=Base Models (LLM-as-a-Judge)2026.01 | 80.09 | |
| URM-LLaMa-3.1-8BModel Category=Scalar Reward Models, Backbone=Llama-3.1-8B2026.01 | 80.04 | |
| Skywork-Reward-Llama-3.1-8B-v0.2Model Category=Scalar Reward Models, Backbone=Llama-3.1-8B2026.01 | 79.07 | |
| RM-NLHF-Qwen-32BModel Category=Our Generative Reward Models, Backbone=Qwen-32B2026.01 | 78.67 | |
| qwen-plus-latestModel Category=Base Models (LLM-as-a-Judge)2026.01 | 77.17 | |
| claude-3-7-sonnet-20250219Model Category=Base Models (LLM-as-a-Judge)2026.01 | 77.14 | |
| RRM-Qwen-32BModel Category=Specialized Generative Reward Models, Backbone=Qwen-32B2026.01 | 73.4 | |
| deepseek-r1-0528Model Category=Base Models (LLM-as-a-Judge)2026.01 | 72.92 | |
| RM-R1-Qwen-32BModel Category=Specialized Generative Reward Models, Backbone=Qwen-32B2026.01 | 72.6 | |
| qwen3-maxModel Category=Base Models (LLM-as-a-Judge)2026.01 | 71.69 | |
| gemini-2.5-proModel Category=Base Models (LLM-as-a-Judge)2026.01 | 71.17 | |
| ArmoRM-Llama3-8B-v0.1Model Category=Scalar Reward Models, Backbone=Llama3-8B2026.01 | 68.97 | |
| gpt-4o-latestModel Category=Base Models (LLM-as-a-Judge)2026.01 | 64.78 | |
| R1-Distill-Qwen-32BModel Category=Base Models (LLM-as-a-Judge), Backbone=Qwen-32B2026.01 | 61.54 | |
| RM-NLHF-Qwen-7BModel Category=Our Generative Reward Models, Backbone=Qwen-7B2026.01 | 57.57 | |
| deepseek-v3Model Category=Base Models (LLM-as-a-Judge)2026.01 | 54.36 | |
| RRM-Qwen-7BModel Category=Specialized Generative Reward Models, Backbone=Qwen-7B2026.01 | 50.82 | |
| RM-R1-Qwen-7BModel Category=Specialized Generative Reward Models, Backbone=Qwen-7B2026.01 | 46.79 | |
| R1-Distill-Llama-8BModel Category=Base Models (LLM-as-a-Judge), Backbone=Llama-8B2026.01 | 40.95 | |
| DeepSeek-R1-Distill-Qwen-7BModel Category=Base Models (LLM-as-a-Judge), Backbone=Qwen-7B2026.01 | 30.52 |