Skip to content

Error while Installing modules for GKD example. #8257

@jeff4700

Description

@jeff4700

Checklist / 检查清单

  • I have searched existing issues, and this is a new question or discussion topic. / 我已经搜索过现有的 issues,确认这是一个新的问题与讨论。

Question Description / 问题描述

Flash Attention Installation Error

git clone https://github.com/modelscope/ms-swift.git
cd ms-swift
uv venv --python python3.12
source .venv/bin/activate
uv pip install -e .
uv pip install flash-attn --no-build-isolation

bash ms-swift/examples/train/rlhf/gkd/think_model.sh

Error:
ImportError: .../flash_attn_2_cuda.cpython-312-x86_64-linux-gnu.so: undefined symbol: _ZN3c104cuda29c10_cuda_check_implementationEiPKcS2_ib


nvidia-smi

Mon Mar  9 20:51:33 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 575.64.03              Driver Version: 575.64.03      CUDA Version: 12.9   

nvcc --version

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2025 NVIDIA Corporation
Built on Tue_May_27_02:21:03_PDT_2025
Cuda compilation tools, release 12.9, V12.9.86
Build cuda_12.9.r12.9/compiler.36037853_0

python -c "import torch; print(torch.version, torch.version.cuda)"
2.10.0+cu128 12.8

What I did:

Reinstalled Cuda toolkit

sudo apt update
sudo apt install -y cuda-toolkit-12-9

sudo ln -sfn /usr/local/cuda-12.9 /usr/local/cuda

Reinstalled pytorch for 12.9

uv pip uninstall torch torchvision torchaudio
uv pip install torch torchvision torchaudio \
  --index-url https://download.pytorch.org/whl/cu129
uv pip install ninja packaging wheel

Then when I run this command, it never proceed for 30 mins.
uv pip install flash-attn --no-build-isolation --no-cache

What should I do?

Metadata

Metadata

Assignees

No one assigned

    Labels

    questionFurther information is requested

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions