UnifiedReward-2.0-qwen35-27b

Model Summary

UnifiedReward-2.0-qwen35-27b is the first unified reward model based on Qwen/Qwen3.5-27B for multimodal understanding and generation assessment, enabling both pairwise ranking and pointwise scoring, which can be employed for vision model preference alignment.

For further details, please refer to the following resources:

• 📰 Paper: https://arxiv.org/pdf/2503.05236
• 🪐 Project Page: https://codegoat24.github.io/UnifiedReward/
• 🤗 Model Collections: https://huggingface.co/collections/CodeGoat24/unifiedreward-models-67c3008148c3a380d15ac63a
• 🤗 Dataset Collections: https://huggingface.co/collections/CodeGoat24/unifiedreward-training-data-67c300d4fd5eff00fa7f1ede
• 👋 Point of Contact: Yibin Wang

vLLM Server Deployment

export VLLM_DISABLE_FLASHINFER_GDN_PREFILL=1
export TOKENIZERS_PARALLELISM=false
vllm serve CodeGoat24/UnifiedReward-2.0-qwen35-27b \
 --host localhost \
 --port 8080 \
 --trust-remote-code \
 --served-model-name UnifiedReward \
 --gpu-memory-utilization 0.95 \
 --mm-encoder-tp-mode data \
 --mm-processor-cache-type shm \
 --enable-prefix-caching \
 --tensor-parallel-size 8 \
 --default-chat-template-kwargs '{"enable_thinking": false}'

The inference code is provided here.

🏁 Compared with Current Reward Models

Citation

@article{unifiedreward,
  title={Unified reward model for multimodal understanding and generation},
  author={Wang, Yibin and Zang, Yuhang and Li, Hao and Jin, Cheng and Wang, Jiaqi},
  journal={arXiv preprint arXiv:2503.05236},
  year={2025}
}