Getting Started
Welcome to Qwodel — production-grade model quantization for Python.
This section will get you from zero to a quantized model in minutes.
In this section
| Page | What you'll learn |
|---|---|
| Installation | How to install Qwodel and the backend extras you need |
| Quick Start | Your first quantization in 5 minutes |
What is Qwodel?
Qwodel is a unified Python library for quantizing large language models (LLMs). It wraps three battle-tested quantization backends behind one clean API:
| Backend | Hardware | Best for |
|---|---|---|
| GGUF | CPU | Broad compatibility, llama.cpp runtimes |
| AWQ | NVIDIA GPU | Maximum GPU inference speed |
| CoreML | Apple Silicon | iOS, macOS, iPadOS deployment |
Next: Installation →