-
Notifications
You must be signed in to change notification settings - Fork 250
Use int4mm weight packing MPS kernel #866
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.
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
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchchat/866
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 0131b58 with merge base d71783c ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Heads up that I have some PR's that'll land to fix most of these test (CI is just broke broke atm) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, but you'll need to update pytorch pin (otherwise your changes would not be visible)
so int4mm path for mps is enabled now? |
* Use int4mm weight packing mps kernel * update torch nightly Updating torch nightly to pick up int4mm groupsize 256 fix (#874)
* Use int4mm weight packing mps kernel * update torch nightly
* Use int4mm weight packing mps kernel * update torch nightly
* Use int4mm weight packing mps kernel * update torch nightly
* Use int4mm weight packing mps kernel * update torch nightly Updating torch nightly to pick up int4mm groupsize 256 fix (#874)
* Use int4mm weight packing mps kernel * update torch nightly Updating torch nightly to pick up int4mm groupsize 256 fix (#874)
This should land after the following PyTorch PR lands: pytorch/pytorch#128965
in order to use the int4mm weight packing MPS kernel introduced there