1、Docker、Docker-Compose安装
https://www.cnblogs.com/a120608yby/p/9883175.html https://www.cnblogs.com/a120608yby/p/14582853.html
2、安装NVIDIA Container Toolkit
https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html
3、基于Docker-Compose部署Xinference
# vim /data/docker/docker-xinference/docker-compose.yml
services:
xinference:
image: xprobe/xinference:latest
container_name: xinference
ports:
- "9997:9997"
restart: unless-stopped
networks:
- ops_default
deploy:
resources:
reservations:
devices:
- driver: nvidia
capabilities: [gpu]
count: all
environment:
- XINFERENCE_HOME=/data
volumes:
- ./data:/data
command: xinference-local -H 0.0.0.0
networks:
ops_default:
external: false
4、启动服务
docker compose up -d
5、访问
访问UI: http://部署主机IP:9997/ui 访问API文档: http://部署主机IP:9997/docs
参考:
https://inference.readthedocs.io/zh-cn/latest/getting_started/
浙公网安备 33010602011771号