export VLLM_DISABLE_FLASHINFER_GDN_PREFILL=1
export TOKENIZERS_PARALLELISM=false
vllm serve CodeGoat24/UnifiedReward-Edit-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}'

@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}
}