Fast Hadamard transform in CUDA, with a PyTorch interface

An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs

llama.cpp fork with additional SOTA quants and improved performance

Built-in Kernel for WSL2

Xanmod kernel for WSL2, built by clang with ThinLTO enabled. Build & Release are automated by Github Action.

A collection of tricks and tools to speed up transformer models

[DEPRECATED] LLR2 is a primality testing program for numbers of several specific forms.

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Code for Deep RL from Human Preferences [Christiano et al]. Plus a webapp for collecting human feedback

Tutorials and implementations for "Self-normalizing networks"