Add initial support for compressed-tensors checkpoints #2732
Conversation
This file contains 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
danieldk
force-pushed
the
feature/compressed-tensors
branch
from
f4fd8d1 to
cd4f20e
Compare
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
danieldk
force-pushed
the
feature/compressed-tensors
branch
2 times, most recently
from
8311452 to
ff9868b
Compare
compressed-tensors is a safetensors extension for sparse, quantized tensors. The format is more powerful than earlier AWQ/GPTQ/FP8 quantization, because - Different quantizer configurations can be used for different targets. - The format can specify input/output quantizers in addition to weight quantizers. - Configurable exclusions for quantization. This change adds a dependency on the `compressed-tensors` package for its configuration parsing and layer matching functionality. The following types of quantization are supported in this PR: - W8A16 and W4A16 INT using GPTQ-Marlin kernels. - W8A8 and W8A16 FP using FP8-Marlin and cutlass kernels. Support for other quantization types will be added in subsequent PRs.
danieldk
force-pushed
the
feature/compressed-tensors
branch
from
fc229c5 to
575af7e
Compare
Narsil
approved these changes
Nov 9, 2024
There was a problem hiding this comment.
LGTM.
The loader logic looks quite complicated (yet again) and seems quite close to existing code.
However I do not have a good solution and the current code is probably close to best in kind (I guess I'm just complaining about the apparently absurd complexity of quantization schemes).
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.
What does this PR do?
compressed-tensors is a safetensors extension for sparse, quantized tensors. The format is more powerful than earlier AWQ/GPTQ/FP8 quantization, because
This change adds a dependency on the
compressed-tensorspackage for its configuration parsing and layer matching functionality.The following types of quantization are supported in this PR:
Support for other quantization types will be added in subsequent PRs.
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.