LLM inference in C/C++

🚀 Efficient implementations for emerging model architectures

A PyTorch native platform for training generative AI models

Data manipulation and transformation for audio signal processing, powered by PyTorch

Rich is a Python library for rich text and beautiful formatting in the terminal.

Development repository for the Triton language and compiler

You like pytorch? You like micrograd? You love tinygrad! ❤️

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Build system, successor to Buck

Kernels, of the mega variety :)

Fast and accurate automatic speech recognition (ASR) for edge devices

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Update ASR paper everyday

Flax is a neural network library for JAX that is designed for flexibility.

Mamba SSM architecture

Interactive visualizations of the geometric intuition behind diffusion models.

Training PyTorch models with differential privacy

Accessible large language models via k-bit quantization for PyTorch.

Tensors, for human consumption

Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry lead…

Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)

A Quirky Assortment of CuTe Kernels

End-to-End Speech Processing Toolkit

Fast and memory-efficient exact attention

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

Tile primitives for speedy kernels

Tools for handling multimodal data in machine learning projects.

Puffing up reinforcement learning

A PyTorch-based Speech Toolkit