A personal AI on machines you own — any agent, loops that run overnight, boards, memory, and git control. Client/server split, always-on, open source end to end.

Enchanted is iOS and macOS app for chatting with private self hosted language models such as Llama2, Mistral or Vicuna using Ollama.

What if you had all the data in the world?

Native iOS and Android client for Open WebUI, plus direct OpenAI-compatible, Ollama, and OpenRouter connections and self-hosted Hermes agents.

Tesseract Open Source OCR Engine (main repository)

Train to 94% on CIFAR-10 in <6.3 seconds on a single A100. Or ~95.79% in ~110 seconds (or less!)

OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.

Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

LlamaIndex is the leading document agent and OCR platform

Cramming the training of a (BERT-type) language model into limited compute.

Hackable and optimized Transformers building blocks, supporting a composable construction.

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

Official implementation of VQ-Diffusion

Development repository for the Triton language and compiler

DALL·E Mini - Generate images from a text prompt

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

Fast and memory-efficient exact attention

Repository for the book

Version control for machine learning

Taming Transformers for High-Resolution Image Synthesis

[NeurIPS‘2021] "TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up", Yifan Jiang, Shiyu Chang, Zhangyang Wang

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

End-to-End Object Detection with Transformers

🤗 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.

The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

Xv6 for RISC-V

Master programming by recreating your favorite technologies from scratch.

A python script to help manage a Gmail inbox by filtering out promotional emails using GPT-3 or GPT-4.