Optimum-quanto

Latest version: v0.2.7

Safety actively analyzes 714860 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 3 of 4

0.0.10

New features:

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

0.0.9

New features:

- quantize weights and activations parameters
- float8 activations

0.0.8

New features:

- weight-only quantization,
- integer matmul acceleration on CUDA.

Bug fixes:

- actually use float16 weights,
- avoid float16 overflows,
- correct device placement,
- robust serialization.

0.0.7

New features:

- per-axis quantization

0.0.6

New features:

- support `opt` models,
- support `gpt-neox` models,
- support `codegen` models.

0.0.5

New features:

- support MPS devices,
- support Transformer models

Page 3 of 4

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.