Stars
Long Context Pre-Training with Lighthouse Attention
SGLang is a high-performance serving framework for large language models and multimodal models.
100M tokens. Infinite compute. Lowest val loss wins.
🐳 Matrix (An open network for secure, decentralized communication) server setup using Ansible and Docker
3D structure generation in Minecraft, powered by fal.ai
A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations
In this repository, 16 models compete to outperform each other in the game Town of Salem. Each model is randomly assigned roles like Vampire, Peasant, Clown, or special variants of Peasants with un…
An open infrastructure to democratize and decentralize the development of superintelligence for humanity.
A collaborative web application (virtual office) presented as a 16-bit RPG video game
A TTS model capable of generating ultra-realistic dialogue in one pass.
Code for "SoftSort: A Continuous Relaxation for the argsort Operator", ICML 2020.
EvaByte: Efficient Byte-level Language Models at Scale
Official Implementation of VCISR: Blind Single Image Super-Resolution with Video Compression Synthetic Data (WACV 2024)
[NeurIPS 2023] Official Code for CycleNet: Rethinking Cycle Consistent in Text‑Guided Diffusion for Image Manipulation
LlamaIndex is the leading document agent and OCR platform
A 15TB Collection of Physics Simulation Datasets
NRL Flux-corrected Algorithm for Solving Generalized Continuity Equations LCPFCT from Python
Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.
Demonstrations of Loss of Plasticity and Implementation of Continual Backpropagation
Waving Goodbye to Low-Res: A Diffusion-Wavelet Approach for Image Super-Resolution
high-accuracy segmentation for anime character
GPU implementation of a fast generalized ANS (asymmetric numeral system) entropy encoder and decoder, with extensions for lossless compression of numerical and other data types in HPC/ML applications.
Officail Implementation for "ReNoise: Real Image Inversion Through Iterative Noising"