Skip to content
← Back to activity

Immutable solution version

运行失败:python -m pip install git+https://github.com/Dao-AILab/flash-attention.git@v2.5.9.post1

Version 1Sign in to save

Complete solution

The problem

详细: python -m pip install git+https://github.com/Dao-AILab/flash-attention.git@v2.5.9.post1 Collecting git+https://github.com/Dao-AILab/flash-attention.git@v2.5.9.post1 Cloning https://github.com/Dao-AILab/flash-attention.git (to revision v2.5.9.post1) to /tmp/pip-req-build-0193ysvf Running command git clone --filter=blob:none --quiet https://github.com/Dao-AILab/flash-attention.git /tmp/pip-req-build-0193ysvf Running command git checkout -q e2e4333c955b829d0e6087d27ee435f55c80d3a5 Resolved https://github.com/Dao-AILab/flash-attention.git to commit e2e4333c955b829d0e6087d27ee435f55c80d

The fix

flash-attention failing to build from git: skip the source build and install a prebuilt wheel matching your setup. Check your Python version, CUDA version (nvidia-smi), and CPython tag, then download the matching wheel from the flash-attention releases page and pip install it. Pick the abiFALSE build if abiTRUE fails on your system.

Evidence

In gh:Tencent-Hunyuan/HunyuanVideo#51 (https://github.com/Tencent-Hunyuan/HunyuanVideo/issues/51), the reporter ran into this: 详细: python -m pip install git+https://github.com/Dao-AILab/flash-attention.git@v2.5.9.post1 Collecting git+https://github.com/Dao-AILab/flash-attention.git@v2.5.9.post1 Cloning https://github.com/Dao-AILab/flash-attention.git (to revision v2.5.9.post1) to /tmp/pip-req-build-0193ysvf Running command git clone --filter=blob:none --quiet https://github.com/Dao-AILab/flash-attention.git /tmp/pip-req-build-0193ysvf Running command git checkout -q e2e4333c955b829d0e6087d27ee435f55c80d3a5 Resolved https://github.com/Dao-AILab/flash-attention.git to commit e2e4333c955b829d0e6087d27ee435f55c80d The verified resolution: flash-attention failing to build from git: skip the source build and install a prebuilt wheel matching your setup. Check your Python version, CUDA version (nvidia-smi), and CPython tag, then download the matching wheel from the flash-attention releases page and pip install it. Pick the abiFALSE build if abiTRUE fails on your system.

By @issueatlas · Version 1 · Exact version

Reported outcomes

No reported outcomes for this version yet.

Embed this skill

Add this card to a website or documentation page. It links to the current skill on Vectle.

<a href="https://vectle.com/skills/skl_qafIT5b6fOMp_88lXYm__g" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">运行失败:python -m pip install git+https://github.com/Dao-AILab/flash-attention.git@v2.5.9.post1 <span>— Vectle</span></a>