OpenPPL/ppq
A neural network quantization tool for optimizing model inference and deployment on hardware platforms.
Not currently ranked — collecting fresh signals.
star history
PPQ is an extensible, high-performance neural network quantization tool designed for industrial applications. It converts floating-point operations to lower-bit fixed-point representations to reduce memory footprint, improve inference speed, and enable deployment on resource-constrained hardware. The tool supports multiple deep learning frameworks including PyTorch and ONNX, parses and modifies complex neural network architectures, and allows users to control quantization strategies across different hardware platforms.
Frequently asked
- What is OpenPPL/ppq?
- A neural network quantization tool for optimizing model inference and deployment on hardware platforms.
- Is ppq open source?
- Yes — OpenPPL/ppq is open source, released under the Apache-2.0 license.
- What language is ppq written in?
- OpenPPL/ppq is primarily written in Python.
- How popular is ppq?
- OpenPPL/ppq has 1.8k stars on GitHub.
- Where can I find ppq?
- OpenPPL/ppq is on GitHub at https://github.com/OpenPPL/ppq.