LLM inference in C/C++

A high-throughput and memory-efficient inference and serving engine for LLMs

Fine-tune mistral-7B on 3090s, a100s, h100s

Finetune Llama 3.2, Mistral, Phi, Qwen 2.5 & Gemma LLMs 2-5x faster with 80% less memory

JAX implementation ViT-VQGAN

DALL·E Mini - Generate images from a text prompt

min(DALL·E) is a fast, minimal port of DALL·E Mini to PyTorch

Fast math typesetting for the web.

scikit-learn: machine learning in Python

JavaScript 3D Library.