An image-to-world skillset for Claude.

Minimal Claude Code alternative. Single Python file, zero dependencies, ~250 lines.

✨ An advanced 3D Gaussian Splatting renderer for THREE.js

Improved sampling via learned diffusions (ICLR2024) and an optimal control perspective on diffusion-based generative modeling (TMLR2024)

Depth Pro: Sharp Monocular Metric Depth in Less Than a Second.

Open Lakehouse Format for Multimodal AI. Convert from Parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, a…

Official inference repo for FLUX.1 models

Utilities intended for use with Llama models.

Cambrian-1 is a family of multimodal LLMs with a vision-centric design.

Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.

llama3 implementation one matrix multiplication at a time

The official Meta Llama 3 GitHub site

Lightplane implements a highly memory-efficient differentiable radiance field renderer, and a module for unprojecting features from images to 3D grids.

A PyTorch native platform for training generative AI models

Easily convert common crawl to a dataset of caption and document. Image/text Audio/text Video/text, ...

Code base of SynthCLIP: CLIP training with purely synthetic text-image pairs from LLMs and TTIs.

NeurIPS 2025 Spotlight; ICLR2024 Spotlight; CVPR 2024; EMNLP 2024

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

Fast bare-bones BPE for modern tokenizer training

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

WIT (Wikipedia-based Image Text) Dataset is a large multimodal multilingual dataset comprising 37M+ image-text sets with 11M+ unique images across 100+ languages.

This repository provides the code and model checkpoints for AIMv1 and AIMv2 research projects.

Building blocks for foundation models.

MLX: An array framework for Apple silicon

A batched offline inference oriented version of segment-anything

Code for the paper "Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models"

The Modular Platform (includes MAX & Mojo)

Fast Implementation of Generalised Geodesic Distance Transform for CPU (OpenMP) and GPU (CUDA)

Code for the paper "Hyperbolic Image-Text Representations", Desai et al, ICML 2023