New Features:
QuantizationModifier
implemented to use compressed-tensors as a backend. (#2307)- INT4 and grouped quantization support (#2307)
Changes:
- UX updated for GPTQModifier. (#2263)
- Upgraded pydantic 1.x to 2.x for compatibility to external dependencies, such as Transformers (#2248)
Resolved Issues:
- None
Known Issues:
- ONNX Export and computer vision models are not officially supported in version 1.8; refer to version v1.7.0 for computer vision support.