100M tokens. Infinite compute. Lowest val loss wins.

Coding Agent singularly focused efficiency and context curation. Reduces API costs by 50-80% vs other agent AND improves the code quality at the same time. Uses Hash Anchored edits, massively paral…

A project that optimizes Whisper for low latency inference using NVIDIA TensorRT

AI agents running research on single-GPU nanochat training automatically

Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with C…

Code for the ShapeR research paper

PyTorch native quantization and sparsity for training and inference

Code release for "Omni3D A Large Benchmark and Model for 3D Object Detection in the Wild"

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

pytorch implementation of VAE-Gumble-Softmax

Code for the paper DisCo-Diff: Enhancing Continuous Diffusion Models with Discrete Latents, ICML 2024

Bayesian entropy estimation in Python - via the Nemenman-Schafee-Bialek algorithm

EDGE: Scalable and optimum mutual information estimator for high-dimensional applications including deep learning

A toolkit for developing and comparing reinforcement learning algorithms.

Minimal implementation of scalable rectified flow transformers, based on SD3's approach

FlexTok: Resampling Images into 1D Token Sequences of Flexible Length

[CVPR'25 Highlight] Official repository of Sonata: Self-Supervised Learning of Reliable Point Representations

Simulation platform for general-purpose robotics & embodied AI learning.

YACS -- Yet Another Configuration System

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

Welcome to the Llama Cookbook! This is your go to guide for Building with Llama: Getting started with Inference, Fine-Tuning, RAG. We also show you how to solve end to end problems using Llama mode…

Implementation of MagViT2 Tokenizer in Pytorch

🌊 [ECCV'24 Oral] MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images

Differentiable gaussian rasterization with depth, alpha, normal map and extra per-Gaussian attributes, also support camera pose gradient

This is the official release for the paper "EFM3D A Benchmark for Measuring Progress Towards 3D Egocentric Foundation Models" (https//arxiv.org/abs/2406.10224).

This is the official repository for "EgoLifter Open-world 3D Segmentation for Egocentric Perception, ECCV 2024"

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