Repository navigation
Initial ParetoQ commit - #1876
Merged
Merged
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1876
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 24191f4 with merge base 6726b0b ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
andrewor14
marked this pull request as draft
March 12, 2025 20:14
andrewor14
added
the
topic: new feature
andrewor14
marked this pull request as ready for review
March 13, 2025 20:31
andrewor14
force-pushed
the
paretoq
branch
2 times, most recently
from
March 14, 2025 16:16
29400c6 to
77b1bcc
Compare
Contributor
|
should there be a test of some sort? Otherwise it's likely this will break soon without anyone knowing. |
andrewor14
force-pushed
the
paretoq
branch
2 times, most recently
from
April 9, 2025 16:05
ca0fdaa to
87638de
Compare
Contributor
Author
added |
jerryzh168
approved these changes
Apr 9, 2025
This project contains the training code of ParetoQ introduced in: "ParetoQ: Scaling Laws in Extremely Low-bit LLM Quantization" (https://arxiv.org/abs/2502.02631). All code is written by @liuzechun and @zxdmike and migrated from https://github.com/facebookresearch/ParetoQ. ParetoQ is the first unified framework that facilitates rigorous comparisons across 1-bit, 1.58-bit, 2-bit, 3-bit, and 4-bit quantization settings. By optimizing training schemes and refining quantization functions, ParetoQ surpasses all previous methods tailored to specific bit widths. Specifically, the 1.58-bit ParetoQ LLaMA-3 8B model reduces the performance gap to full precision by relatively 37.8% compared to the 1-bit Era’s 1.58-bit LLaMA-3 8B model, while using only 30% of the training tokens.
liangel-02
pushed a commit
that referenced
this pull request
Aug 25, 2025
This project contains the training code of ParetoQ introduced in: "ParetoQ: Scaling Laws in Extremely Low-bit LLM Quantization" (https://arxiv.org/abs/2502.02631). All code is written by @liuzechun and @zxdmike and migrated from https://github.com/facebookresearch/ParetoQ. ParetoQ is the first unified framework that facilitates rigorous comparisons across 1-bit, 1.58-bit, 2-bit, 3-bit, and 4-bit quantization settings. By optimizing training schemes and refining quantization functions, ParetoQ surpasses all previous methods tailored to specific bit widths. Specifically, the 1.58-bit ParetoQ LLaMA-3 8B model reduces the performance gap to full precision by relatively 37.8% compared to the 1-bit Era’s 1.58-bit LLaMA-3 8B model, while using only 30% of the training tokens.
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
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 project contains the training code of ParetoQ introduced in: "ParetoQ: Scaling Laws in Extremely Low-bit LLM Quantization" (https://arxiv.org/abs/2502.02631). All code is written by @liuzechun and @zxdmike and migrated from
https://github.com/facebookresearch/ParetoQ.
ParetoQ is the first unified framework that facilitates rigorous comparisons across 1-bit, 1.58-bit, 2-bit, 3-bit, and 4-bit quantization settings. By optimizing training schemes and refining quantization functions, ParetoQ surpasses all previous methods tailored to specific bit widths. Specifically, the 1.58-bit ParetoQ LLaMA-3 8B model reduces the performance gap to full precision by relatively 37.8% compared to the 1-bit Era’s 1.58-bit LLaMA-3 8B model, while using only 30% of the training tokens.