[fp8] Only assert when CUDA is available. #2590
Draft
+45
−37
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2590
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
Stonepia
changed the title
liangan1
suggested changes
Jul 24, 2025
|
|
||
| assert is_sm_at_least_89() or is_MI300(), ( | ||
| "Float8 dynamic quantization requires CUDA compute capability ≥8.9 or MI300+." | ||
| ) |
There was a problem hiding this comment.
This change will impact other device, such as cpu or any unknow device. Suggest to add common utlis function to judge the fp8 capability and apply it to all fp8 related changes. This function should only ensure the CUDA compute capability ≥8.9 or MI300+ and XPU device is available now.
This commit performs the capability checks only when CUDA is available, so that we have better support for third-party devices like Intel GPU.
|
Can you show the accuracy result for dq-fp8 for both CUDA and XPU based on the LLama-3.1-8B? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This commit performs the capability checks only when CUDA is available, so that we have better support for third-party devices like Intel GPU.