[Release_v2160] Update Release notes #3380
Open
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
|
@alexsu52, @ljaljushkin, @l-bat, @nikita-savelyevv, @andreyanufr, @andrey-churkin, @daniil-lyakhov, @kshpv, @AlexanderDokuchaev, @anzr299 fill the document with your changes for the upcoming release, please. |
github-actions
bot
added
documentation
MaximProshin
requested review from
alexsu52,
ljaljushkin,
l-bat,
nikita-savelyevv,
kshpv,
anzr299,
AlexanderDokuchaev,
andrey-churkin,
andreyanufr,
daniil-lyakhov and
MaximProshin
nikita-savelyevv
approved these changes
Mar 27, 2025
| - Features: | ||
| - ... | ||
| - Fixes: | ||
| - Fixed occasional failures of weight compression algorithm on ARM CPUs. |
| - Fixes: | ||
| - Fixed occasional failures of weight compression algorithm on ARM CPUs. | ||
| - Improvements: | ||
| - Reduced the run time and peak memory of mixed precision assignment procedure during weight compression in the OpenVINO backend. Overall compression time reduction in mixed precision case is about 20-40%; peak memory reduction is about 20%. |
ljaljushkin
reviewed
Mar 27, 2025
ReleaseNotes.md
Outdated
| - General: | ||
| - ... | ||
| - Features: | ||
| - (Torch) Introduced a novel weight compression method for Large Language Models (LLMs) that significantly improves accuracy with int4 weights. Leveraging Quantization-Aware Training (QAT) and absorbable LoRA adapters, this approach can achieve a 2x reduction in accuracy loss during compression compared to the best post-training weight compression technique in NNCF (Scale Estimation + AWQ + GPTQ). The `nncf.compress_weight` API now includes a new `compression_format` option, `CompressionFormat.FQ_LORA`, for this QAT method, and a sample compression pipeline with preview support is available [here](examples/llm_compression/torch/qat_with_lora). |
ljaljushkin
reviewed
Mar 27, 2025
| - (Torch) Introduced a novel weight compression method for Large Language Models (LLMs) that significantly improves accuracy with int4 weights. Leveraging Quantization-Aware Training (QAT) and absorbable LoRA adapters, this approach can achieve a 2x reduction in accuracy loss during compression compared to the best post-training weight compression technique in NNCF (Scale Estimation + AWQ + GPTQ). The `nncf.compress_weight` API now includes a new `compression_format` option, `CompressionFormat.FQ_LORA`, for this QAT method, and a sample compression pipeline with preview support is available [here](examples/llm_compression/torch/qat_with_lora). | ||
| - Fixes: | ||
| - Fixed occasional failures of weight compression algorithm on ARM CPUs. | ||
| - (Torch) Fixed weight compression for float16/bfloat16 models. |
reworked FQ + Lora
ljaljushkin
approved these changes
Mar 27, 2025
anzr299
reviewed
Mar 28, 2025
|
|
||
| Requirements: | ||
|
|
||
| - Updated PyTorch (2.6.0) and Torchvision (0.21.0) versions. |
| - (Torch) Fixed weight compression for float16/bfloat16 models. | ||
| - Improvements: | ||
| - Reduced the run time and peak memory of mixed precision assignment procedure during weight compression in the OpenVINO backend. Overall compression time reduction in mixed precision case is about 20-40%; peak memory reduction is about 20%. | ||
| - (TorchFX, Experimental) Added quantization support for (TorchFX)[https://pytorch.org/docs/stable/fx.html] models exported with dynamic shapes. |
anzr299
approved these changes
Mar 28, 2025
kshpv
reviewed
Mar 28, 2025
| - ... | ||
| - Features: | ||
| - (Torch) Introduced a novel weight compression method to significantly improve the accuracy of Large Language Models (LLMs) with int4 weights. Leveraging Quantization-Aware Training (QAT) and absorbable LoRA adapters, this approach can achieve a 2x reduction in accuracy loss during compression compared to the best post-training weight compression technique in NNCF (Scale Estimation + AWQ + GPTQ). The `nncf.compress_weight` API now includes a new `compression_format` option, `CompressionFormat.FQ_LORA`, for this QAT method, and a sample compression pipeline with preview support is available [here](examples/llm_compression/torch/qat_with_lora). | ||
| - (Torch) Add support for 4-bit weight compression, along with AWQ and Scale Estimation data-aware methods to reduce quality loss after compression. |
kshpv
approved these changes
Mar 28, 2025
l-bat
approved these changes
Mar 28, 2025
andreyanufr
approved these changes
Mar 28, 2025
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.
Changes
Reason for changes
Related tickets
For the contributors:
Please add your changes (as a commit to the branch) to the list according to the template and previous notes
Do not add tests-related notes
Provide the list of the PRs (for all your notes) in the comment for the discussion