-
Notifications
You must be signed in to change notification settings - Fork 12
Introduction
leafspark edited this page Aug 31, 2024
·
2 revisions
AutoGGUF is an automated graphical interface designed for GGUF model quantization. Made with PySide6 and llama.cpp, AutoGGUF simplifies the process of quantizing large language models (LLMs) for efficient local inference.
The primary purpose of AutoGGUF is to democratize access to LLMs by making quantization more accessible. Key features include:
- Automated download and management of llama.cpp backends (including CUDA support)
- Easy model selection and quantization
- Support for various quantization types
- User-friendly graphical interface
- Parallel quantization
- LoRA conversion