Download Latest Version
v3.2.0 source code.zip (6.5 MB)
Get an email when there's a new version of NNCF
| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.md | 2026-06-01 | 3.9 kB | |
| v3.2.0 source code.tar.gz | 2026-06-01 | 5.6 MB | |
| v3.2.0 source code.zip | 2026-06-01 | 6.5 MB | |
| Totals: 3 Items | 12.1 MB | 0 | |
- Features:
- Add support for ignoring specific weight nodes by constant node name within
IgnoredScopefornncf.compress_weightsandnncf.prune(https://github.com/openvinotoolkit/nncf/pull/4063) (https://github.com/openvinotoolkit/nncf/pull/4064). - Fixes:
- (PyTorch) Fixed TorchFX model extractor to prevent capture of nodes that do not affect target outputs (https://github.com/openvinotoolkit/nncf/pull/3982).
- Fixed
QuantizationConstraints.from_config_dictto avoid unexpected behavior (https://github.com/openvinotoolkit/nncf/pull/4046). - (OpenVINO) Fixed Scale Estimation to properly handle zero
initial_stepsandscale_stepsincalculate_quantization_params(https://github.com/openvinotoolkit/nncf/pull/4039). - Improvements:
- (OpenVINO) Added
transpose_aattribute support for Scale Estimation algorithm in ONNX and OpenVINO backends (https://github.com/openvinotoolkit/nncf/pull/3839). - (PyTorch) Removed no-op operations (
to,dropout, etc.) from the inference graph in PTQ and Weight Compression algorithms (https://github.com/openvinotoolkit/nncf/pull/4057). - (PyTorch) Improved
load_from_configreliability by saving module path in serialized config (https://github.com/openvinotoolkit/nncf/pull/4050) (https://github.com/openvinotoolkit/nncf/pull/4054). - Tutorials:
- Post-Training Optimization of Qwen3-VL Embedding and Reranker Models
- Post-Training Optimization of MiniCPM-V 4.6 Model
- Post-Training Optimization of MinerU 2.5 Model
- Post-Training Optimization of GLM-OCR Model
- Post-Training Optimization of FLUX.2 Klein Model
- Post-Training Optimization of MedASR Model
- Post-Training Optimization of ERNIE-Image-Turbo Model
- Post-Training Optimization of VoxCPM2 Model
- Post-Training Optimization of HY-MT1.5 Models
- Post-Training Optimization of YOLO26 Models
- Deprecations/Removals:
- (PyTorch) Removed
example_inputparameter fromload_from_config, which has been unused since 3.0.0 (https://github.com/openvinotoolkit/nncf/pull/4051). - (PyTorch) Removed
OVQuantizerfrom NNCF; use it from theexecutorchpackage instead (https://github.com/openvinotoolkit/nncf/pull/3842). - Removed deprecated
CompressWeightsMode.INT8mode. UseCompressWeightsMode.INT8_SYMorCompressWeightsMode.INT8_ASYMinstead (https://github.com/openvinotoolkit/nncf/pull/4008). - Removed unused
ProgressBarclass (https://github.com/openvinotoolkit/nncf/pull/4044). - Requirements:
- Updated
onnxto 1.21.0 (https://github.com/openvinotoolkit/nncf/pull/4006). - Updated
torchaoto 0.17.0 (https://github.com/openvinotoolkit/nncf/pull/4013). - Updated
onnxscriptto 0.6.2 (https://github.com/openvinotoolkit/nncf/pull/4015).
