-
Notifications
You must be signed in to change notification settings - Fork 569
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
[Docs] Fix A100 Support for DeepSeek-R1 671B: Clarify FP8 Issue & YAML Inconsistencies #4726
Conversation
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.
Thanks @andylizf for updating this ! It looks great. Left some comments. Could formatting the output of the model in the details as well, ie make those \n actual new lines, like we did in the distilled model example readme?
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.
Thanks @andylizf
Co-authored-by: Zhanghao Wu <[email protected]>
Resolves #4723
A100 does not support FP8, but the current example incorrectly lists
A100
as a valid accelerator.This PR removes A100 from the default accelerator list and introduces a dedicated job YAML for running DeepSeek-R1 on A100 with BF16 conversion. It also adds a documentation section explaining how to use A100 properly.
Additionally, this PR addresses several documentation inconsistencies and improves wording for clarity.