Implementation of Alphafold 3 from Google Deepmind in Pytorch

Central repository for biomolecular foundation models with shared trainers and pipeline components

The best ChatGPT that $100 can buy.

Fast and memory-efficient exact attention

A concise but complete full-attention transformer with a set of promising experimental features from various papers

Open-source release accompanying Gao et al. 2025

Code for running RFdiffusion

AlphaFold 3 inference pipeline.

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

[CVPR 2025, Spotlight] SimLingo (CarLLava): Vision-Only Closed-Loop Autonomous Driving with Language-Action Alignment

LLM training in simple, raw C/CUDA

RAG that intelligently adapts to your use case, data, and queries

A simple, easy-to-hack Vector Database

A simple, easy-to-hack GraphRAG implementation

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

Because tinygrad got out of hand with line count

Vector (and Scalar) Quantization, in Pytorch

Implementation of Denoising Diffusion Probabilistic Model in Pytorch

If tinygrad wasn't small enough for you...

Implementing a ChatGPT-like LLM from scratch, step by step

My implementation of Stanford CS336 assignments.

Official repository for the Boltz biomolecular interaction models

The simplest, fastest repository for training/finetuning medium-sized GPTs.