CUDA_VISIBLE_DEVICES=4,5,6,7 python tools/eval_hmmt2025_vllm.py \
  --model /mnt/disk1/yihao/Lightning-OPD/checkpoints/qwen3-4b-lightning-opd-hf \
  --num-gpus 4 \
  --prompt-template paper \
  --n-samples 32 \
  --temperature 0.6 \
  --top-p 0.95 \
  --max-tokens 32768 \
  --enable-thinking \
  --output outputs/hmmt_feb_2025_qwen3_4b_lightning_opd.jsonl