Utilities intended for use with Llama models.

Entropy Based Sampling and Parallel CoT Decoding

A massively parallel, optimal functional runtime in Rust

A massively parallel, high-level programming language

Vision Transformer explanation and implementation with PyTorch

A composable build system for OCaml.

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

The financial transactions database designed for mission critical safety and performance.

Solve puzzles. Improve your pytorch.

[NeurIPS 2023] Text Promptable Surgical Instrument Segmentation with Vision-Language Models

⚡ A Fast, Extensible Progress Bar for Python and CLI

Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

Code release for NeRF (Neural Radiance Fields)

tracking medical datasets, with a focus on medical imaging

Insight Toolkit (ITK) -- Official Repository. ITK builds on a proven, spatially-oriented architecture for processing, segmentation, and registration of scientific images in two, three, or more dime…

The MAX Platform (includes Mojo)

</> htmx - high power tools for HTML

Make your functions return something meaningful, typed, and safe!

Smart, pythonic, ad-hoc, typed polymorphism for Python

Convert Pydantic from V1 to V2 ♻

You like pytorch? You like micrograd? You love tinygrad! ❤️

Inference Llama 2 in one file of pure C

The official PyTorch implementation for NCSNv2 (NeurIPS 2020)

An extremely fast Python linter and code formatter, written in Rust.

Denoising Diffusion Implicit Models

Tensors and Dynamic neural networks in Python with strong GPU acceleration

WebGL-based viewer for volumetric data

Elucidating the Design Space of Diffusion-Based Generative Models (EDM)

Command line client for Nix binary cache hosting:

High-Resolution Image Synthesis with Latent Diffusion Models