haxiv — Deep learning framework for LLMs (Llama/Gemma/Qwen) on CPU, Apple MLX, Metal, CUDA. Load PyTorch/ONNX/TF/GGUF with zero

haxiv is a deep learning framework for LLMs that prioritizes flexibility and broad hardware support. It enables zero-conversion loading of models from PyTorch, ONNX, TensorFlow, and GGUF formats, running them on CPUs, Apple Silicon via MLX and Metal, and CUDA. This makes haxiv a compelling alternative to existing frameworks for developers working with a variety of models and hardware setups.

Opening Kapyn…