Skip to content
/ Fuser Public

A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")

License

Notifications You must be signed in to change notification settings

NVIDIA/Fuser

Folders and files

NameName
Last commit message
Last commit date
Feb 4, 2025
Feb 16, 2025
Feb 20, 2024
Feb 17, 2025
Dec 13, 2024
Nov 7, 2024
Jan 29, 2025
Feb 14, 2025
Feb 3, 2025
Feb 15, 2025
Mar 1, 2024
Feb 4, 2025
Jun 5, 2023
Nov 22, 2024
Mar 13, 2023
Apr 15, 2024
Apr 6, 2023
Jan 29, 2025
Feb 12, 2025
Mar 13, 2023
Nov 21, 2024
Aug 26, 2024
Mar 13, 2023
Apr 8, 2023
Nov 7, 2024
Feb 1, 2025

Repository files navigation

Fuser

A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")

Installation

We publish nightly wheel packages on https://pypi.nvidia.com, while build against stable torch version on https://pypi.org. Wheels are published for Python version: 3.10, 3.12.

built-env cuda 11.8 cuda 12.1 cuda 12.4
torch 2.5 (pypi.org) nvfuser-cu118-torch25 nvfuser-cu121-torch25 nvfuser-cu124-torch25
torch nightly (pypi.nvidia.com) nvfuser-cu118 nvfuser-cu121 nvfuser-cu124

Note that nvfuser built against stable torch version isn't compatible with nightly pytorch wheel, so ensure you pick the right version suiting your environment.

nightly nvfuser pip wheel

You can instll a nightly nvfuser pip package built against torch nightly code base with pip install --pre nvfuser-cu121 --extra-index-url https://pypi.nvidia.com

As we build against nightly torch wheel and there's no compatibility promised on nightly wheels, we have explicitly marked the nightly torch wheel as an optinoal dependency. You can choose to install the torch wheel along with nvfuser package. e.g. pip install --pre "nvfuser-cu121[torch]" --extra-index-url https://pypi.nvidia.com. Note that this may uninstall your local pytorch installation and install the compatible nightly pytorch.

nvfuser pip wheel against pytorch stable release

nvfuser pip wheel built against stable torch releases is published on pypi.org. Pick the right cuda toolkit version to match your torch installation. e.g. pip install nvfuser-cu121-torch24

PyPI: https://pypi.org/project/nvfuser/

Developer

Docs: https://github.com/NVIDIA/Fuser/wiki

Supported compilers:

GCC:

We support all "supported releases" of gcc as specified in the official site. As of 6/20/2024, they are:

  • gcc 11.4
  • gcc 12.4
  • gcc 13.2
  • gcc 14.1

Clang:

  • clang 14+

Supported C++ standard:

  • C++17
  • C++20

We are actively considering dropping C++17 support