rubricreward/R3-Qwen3-14B-14k

rubricreward/R3-Qwen3-14B-14k is a 14 billion parameter reward model from the R3 family, fine-tuned from Qwen/Qwen3-14B. It is specifically designed for robust, rubric-agnostic evaluation across diverse tasks like classification, preference optimization, and question answering. This model excels at providing detailed assessments and scores based on given rubrics and reasoning, making it suitable for automated content evaluation.