RewardAnything-8B-v1-f32-GGUF

RewardAnything-8B-v1 is a generalizable, principle-following reward model with 8B parameters based on Qwen3-8B, designed to interpret and apply natural language principles directly at inference time for dynamic adaptation across diverse evaluation criteria without retraining; it achieves state-of-the-art results on RM-Bench and RABench, demonstrates strong generalization to new, unseen reward principles, supports transparent reasoning to explain its decisions, works efficiently with standard RLHF (PPO, GRPO) pipelines, and offers flexible deployment for local use, batch inference, or direct Hugging Face integration, all under the Apache 2.0 license for both research and production-scale applications.

Execute using Ollama

Model Files

Quants Usage