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

Jump to bottom

INT4 support on Volta? #739

Closed
sleepwalker2017 opened this issue Dec 25, 2023 · 2 comments · Fixed by #787
Closed

INT4 support on Volta? #739

sleepwalker2017 opened this issue Dec 25, 2023 · 2 comments · Fixed by #787
Assignees
@Tracin
Labels
documentation Improvements or additions to documentation triaged Issue has been triaged by maintainers

Comments

@sleepwalker2017
Copy link

image
Seems int4 quantization is not supported on V100, but why the documentation says int4 is supported on Volta?
@juney-nvidia
Copy link
Collaborator

@sleepwalker2017 Thanks for reporting this, this is indeed a typo. And we will fix it soon.

June

@juney-nvidia juney-nvidia added documentation Improvements or additions to documentation triaged Issue has been triaged by maintainers labels Dec 25, 2023
@juney-nvidia
Copy link
Collaborator

The INT4 described here means INT4 weight-only, rather than the INT4 Tensor Core. We will refine the documentation to make it clearer.

Thanks
June

@kaiyux kaiyux mentioned this issue Jan 2, 2024
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@Tracin Tracin

Labels
documentation Improvements or additions to documentation triaged Issue has been triaged by maintainers
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Update TensorRT-LLM NVIDIA/TensorRT-LLM
3 participants
@Tracin @sleepwalker2017 @juney-nvidia