serve_vllm.sh

#!/bin/bash

python -m vllm.entrypoints.openai.api_server \
    --model moonshotai/Kimi-Dev-72B \
    --tensor-parallel-size 4 \
    --max-num-seqs 4 \
    --max-model-len 65536 \
    --gpu-memory-utilization 0.9 \
    --host localhost \
    --served-model-name kimi-dev \
    --port 8080