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Outdated suggestions cannot be applied.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 unobtrusively adds support for AMD devices, in a way that minimizes changes or adding new code.
Performance with bfloat16 on a 7900 XTX is ~50,000 toks/sec for a single GPU, and ~210,000 toks/sec for 4x GPUs (as a frame of reference, the latest pytorch 2.4.0.dev20240513 runs at ~42,000 toks/sec on a single device).
Should this be merged here, or maintained as a separate fork?