Optimum

Latest version: v1.24.0

Safety actively analyzes 723685 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 21 of 23

0.1.1

With this release, we enable Intel Neural Compressor v1.7 PyTorch dynamic, post-training and aware-training quantization for a variety of NLP tasks. This support includes the overall process, from quantization application to the loading of the resulting quantized model. The latter being enabled by the introduction of the `IncQuantizedModel` class.

0.1.0

New features

- group-wise quantization,
- safe serialization.

0.0.13

New features

- new `QConv2d` quantized module,
- official support for `float8` weights.

Bug fixes

- fix `QbitsTensor.to()` that was not moving the inner tensors,
- prevent shallow `QTensor` copies when loading weights that do not move inner tensors.

0.0.12

New features

- quanto kernels library (not used for now in quantize).

Breaking changes

- quantization types are now all quanto.dtype

0.0.11

New features:

- support int2 and int4 weights.

New contributors:

younesbelkada
a-r-r-o-w

0.0.10

New features:

- calibration streamline option to remove spurious quantize/dequantize,
- calibration debug mode.

Page 21 of 23

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.