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-base/flagos-base-enflame-tops1.10.6:2.1.2 open_in_new

Python

3.12

Major Python packages

  • enflame-modelopt==3.6.20260615+torch.2.11.0
  • flag_gems==5.3.4
  • flagtree==0.6.1+enflame3.6
  • flash-attn==2.7.2+torch.2.11.0.gcu.3.8.20260706
  • numpy==2.3.5
  • pyefml==1.10.6
  • tops-extension==3.6.20260529+torch.2.11.0
  • topstx==1.10.6
  • torch-gcu==2.11.0+3.8.20260713
  • torch==2.11.0+cpu
  • torchaudio==2.11.0+cpu
  • torchcodec==3.8.20260624+torch.2.11.0
  • torchvision==0.26.0+cpu
  • triton==3.6.0 (+ triton-gcu==3.6.0+1.0.20260722)

Switch compiler

This image includes both FlagTree (default) and Triton. To switch, run compiler triton inside the container. Use compiler flagtree to switch back, or compiler to check the active compiler.

Environment

  • TORCH_GCU_ENABLE_INT64_AND_UINT64=1
  • ENABLE_I64_CHECK=0
  • TORCHGCU_INDUCTOR_ENABLE=1
  • ENFLAME_PT_OP_DEBUG_CONFIG=fallback_cpu=all_ops

Launch

With the container toolkit (optional):

  docker run --rm -it \
  --network host \
  -e ENFLAME_VISIBLE_DEVICES=all \
  -e TENCENT_VISIBLE_DEVICES=all \
  harbor.baai.ac.cn/flagos-runtime/flagos-runtime-enflame-tops1.10.6:2.1.2 bash

Without a toolkit — plain docker / podman:

  docker run --rm -it \
  --privileged \
  -v /dev:/dev \
  harbor.baai.ac.cn/flagos-runtime/flagos-runtime-enflame-tops1.10.6:2.1.2 bash

Verify

Inside the container, confirm the accelerator is visible:

  efsmi
Edit this page

Last updated 15 Aug 2026, 12:51 +0800. history