model/UniVectorSQL-7B-LoRA/run_vllm.sh

#!/bin/bash
set -euo pipefail

MODEL_DIR="$(cd "$(dirname "$0")" && pwd)"
PORT=8004

python -m vllm.entrypoints.openai.api_server \
  --model "$MODEL_DIR" \
  --tensor-parallel-size 1 \
  --port $PORT \
  --max-num-seqs 1 --gpu-memory-utilization 0.95