Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

bump docs to PyTorch 2.5.1 and corresponding nvfuser #1377

Merged
merged 1 commit into from
Oct 31, 2024
+5 −5
Merged

bump docs to PyTorch 2.5.1 and corresponding nvfuser #1377

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,11 +81,11 @@ pip install git+https://github.com/Lightning-AI/lightning-thunder.git@main

To achieve the best performance, you can install Thunder with the following additional dependencies:

- install nightly [nvFuser](https://github.com/NVIDIA/Fuser) built for PyTorch 2.4 as follows:
- install prerelease [nvFuser](https://github.com/NVIDIA/Fuser) built for PyTorch 2.5.1 as follows:

```bash
# install nvFuser built for the matching stable PyTorch
pip install --pre nvfuser-cu121-torch24
pip install --pre nvfuser-cu121-torch25
```

- install [cudnn](https://gitlab-master.nvidia.com/cudnn/cudnn_frontend) as follows:
Expand Down
6 changes: 3 additions & 3 deletions docs/source/fundamentals/installation.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,12 +6,12 @@ Minimal dependencies

Follow these instructions to install PyTorch, nvFuser, and finally Thunder.

Install PyTorch and nvFuser with pip (command shown is for CUDA 12.1 and PyTorch 2.4.x)::
Install PyTorch and nvFuser with pip (command shown is for CUDA 12.1 and PyTorch 2.5.x)::

pip install --pre nvfuser-cu121-torch24
pip install --pre nvfuser-cu121-torch25

cu121 can be replaced with cu118 depending on your CUDA version. NVFuser builds typically support the latest point release of PyTorch stable versions.
For torch 2.4, cu124 is also supported. For nightly versions and more detailed instructions, please see https://github.com/NVIDIA/Fuser/#installation
For torch 2.5, cu124 is also supported. For nightly versions and more detailed instructions, please see https://github.com/NVIDIA/Fuser/#installation

You're all set with minimal dependencies, so you can follow `Install Thunder`_.

Expand Down
Loading