feat(mps): implement native-like Float8 support via LUT dequantization #11669
+113
−5
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
Add a new MPS-specific operations module to handle Float8 tensor support on Apple Silicon. Since MPS does not natively support Float8 dtypes, this implementation uses a uint8 storage strategy combined with a GPU-accelerated Lookup Table (LUT) for efficient dequantization, keeping data on the GPU. - Add comfy/mps_ops.py: Implement cached LUT generation and index-based dequantization for MPS. - Modify comfy/quant_ops.py: Add logic to view Float8 tensors as uint8 when moving to MPS, and route dequantization to mps_ops. - Modify comfy/float.py: Add CPU staging for stochastic rounding to prevent MPS casting errors during quantization. - Modify comfy/quant_ops.py: Add fallback for fp8_linear. Signed-off-by: Macpaul Lin <[email protected]>
macpaul
requested review from
Kosinkadink,
comfyanonymous and
guill
as code owners
Member
Test Evidence CheckIf this PR modifies behavior that requires testing, a test explanation is required. PRs lacking applicable test explanations may not be reviewed until added. Please add test explanations to ensure code quality and prevent regressions. If this PR changes user-facing behavior, visual proof (screen recording or screenshot) is required. PRs without applicable visual documentation may not be reviewed until provided. You can add it by:
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
Add a new MPS-specific operations module to handle Float8 tensor support on Apple Silicon. Since MPS does not natively support Float8 dtypes, this implementation uses a uint8 storage strategy combined with a GPU-accelerated Lookup Table (LUT) for efficient dequantization, keeping data on the GPU.