[FA3] uv installation support

[bbuschkaemper]

Motivation

FA3, even though in beta, is supported in many downstream frameworks, e.g. megatron-core. To streamline installation using uv and pyproject.toml, the building of the FA3 beta needs to be adapted: Currently, building using uv results in all package files being dropped directly into site-packages/, instead of the corresponding site-packages/flash_attn_3 subfolder.

Existing downstream frameworks like megatron-core need a flash_attn_3 package to work (see here), preventing a clean building solution using a single pyproject.toml and requiring custom build steps.

Solution

To support projects to integrate FA3 correctly, I propose a backwards-compatible change to the package setup, allowing FA3 to be properly built into a package using a pyproject.toml while maintaining existing behaviour.

This change outputs the flash_attn_interface.py and flash_attn_config.py files into both the site-packages/ (for backwards compatibility) and site-packages/flash_attn_3 folders.

[Copilot]

Pull request overview

Adds an opt-in build flag to install FlashAttention-3 (FA3) as a flash_attn_3 package (for uv/pyproject.toml workflows) while keeping the default legacy layout unchanged.

Changes:

• Introduce FLASH_ATTENTION_AS_PACKAGE build flag in hopper/setup.py to install modules under flash_attn_3.* and force source builds (skip cached wheels) when enabled.
• Ensure flash_attn_config.py is generated in AS_PACKAGE scenarios even when CUDA build is skipped.
• Document uv installation configuration in README.md.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
hopper/setup.py	Adds AS_PACKAGE build mode and adjusts wheel build/config generation to support package-style installs.
README.md	Adds uv/pyproject.toml installation instructions using the new build flag.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 5 changed files in this pull request and generated 2 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 5 changed files in this pull request and generated 6 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[bbuschkaemper]

@tridao This PR is supposed to be an easy, backwards-compatible change to allow clean fa3 builds with uv so that packages that require a "flash_attn_3" package instead of a single "flash_attn_interface" module will work with it. If you have a different plan for fa3 building let me know; I can adapt the PR as-needed.