| 1234567891011121314151617 |
- #!/bin/bash
- set -e
- # 创建日志目录
- mkdir -p /var/log/vllm
- # 启动 vLLM 服务
- python3 -m vllm.entrypoints.openai.api_server \
- --model /model/Qwen3-Embedding-8B \
- --runner pooling \
- --convert embed \
- --trust-remote-code \
- --tensor-parallel-size 1 \
- --gpu-memory-utilization 0.45 \
- --host 0.0.0.0 \
- --port 30000 \
- --api-key lq123456 2>&1 | tee /var/log/vllm/qwen3-embedding-8b-server.log
|
