Utilities intended for use with Llama models.

Fast Matrix Multiplications for Lookup Table-Quantized LLMs

Large Language Model Text Generation Inference

Distillation version of llama-68m only for MLsys research use.

Custom data types and layouts for training and inference

Collection of the latest, greatest, deep learning optimizers (for Pytorch) - CNN, NLP suitable

Port of Andrej Karpathy's nanoGPT to Apple MLX framework.

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

LLM101n: Let's build a Storyteller

[ICLR2024 spotlight] OmniQuant is a simple and powerful quantization technique for LLMs.

Prune transformer layers

llama3.cuda is a pure C/CUDA implementation for Llama 3 model.

A native PyTorch Library for large model training

Deep Reinforcement Learning: Zero to Hero!

Extract Deflate64 ZIP archives with Python's `zipfile` API.

The official Meta Llama 3 GitHub site

A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.

LLM training in simple, raw C/CUDA

An introduction to programming in Triton

Applied AI experiments and examples for PyTorch

A Native-PyTorch Library for LLM Fine-tuning

The easiest way to run the fastest MLX-based LLMs locally