Prerequisites

  • Architecture: x86_64
  • Chip models: Enflame Zixiao C200 (S60)
  • Host driver: 1.10.6
  • Container toolkit (optional) : tencent-container-toolkit >= 2.0.52

Image contents

Built on

harbor.baai.ac.cn/flagos-runtime/flagos-runtime-enflame-tops1.10.6:2.2.0 open_in_new

Python

3.12

Application package

vllm==0.24.0+flagos

vllm-plugin-fl==0.3.0

Launch

Published: harbor.baai.ac.cn/flagos-app/vllm0.24.0-enflame-tops1.10.6:2.2.0-0.3.0

The image name is long — assign it to a variable first:

  IMG=harbor.baai.ac.cn/flagos-app/vllm0.24.0-enflame-tops1.10.6:2.2.0-0.3.0

The two approaches below are alternatives — pick the one that matches how your host runs containers:

With the container toolkit

Start an interactive shell:

  docker run --rm -it \
  --network host \
  -e ENFLAME_VISIBLE_DEVICES=all \
  -e TENCENT_VISIBLE_DEVICES=all \
  $IMG bash

Start the app with its default settings:

  docker run --rm -it \
  --network host \
  -e ENFLAME_VISIBLE_DEVICES=all \
  -e TENCENT_VISIBLE_DEVICES=all \
  $IMG

Pass arguments to the launcher:

  docker run --rm -it \
  --network host \
  -e ENFLAME_VISIBLE_DEVICES=all \
  -e TENCENT_VISIBLE_DEVICES=all \
  $IMG vllm-serve --model <path> --port 9000

Without a toolkit — plain docker / podman

Start an interactive shell:

  docker run --rm -it \
  --privileged \
  -v /dev:/dev \
  $IMG bash

Start the app with its default settings:

  docker run --rm -it \
  --privileged \
  -v /dev:/dev \
  $IMG

Pass arguments to the launcher:

  docker run --rm -it \
  --privileged \
  -v /dev:/dev \
  $IMG vllm-serve --model <path> --port 9000
Edit this page

Last updated 28 Sep 2026, 22:20 +0800. history