SGLang is a fast serving framework for large language models and vision language models.

A minimal GPU design in Verilog to learn how GPUs work from the ground up

A generative world for general-purpose robotics & embodied AI learning.

Official implementation for SIGGRAPH 2023 paper "Learning Physically Simulated Tennis Skills from Broadcast Videos"

Fast and memory-efficient exact attention

xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism

Run your own AI cluster at home with everyday devices 📱💻 🖥️⌚

Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and…

A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API

Ongoing research training transformer models at scale

FlagScale is a large model toolkit based on open-sourced projects.

The C++ Core Guidelines are a set of tried-and-true guidelines, rules, and best practices about coding in C++

μ-Cuda, COVER THE LAST MILE OF CUDA. With features: intellisense-friendly, structured launch, automatic cuda graph generation and updating.

Free ChatGPT Site List 这儿为你准备了众多免费好用的ChatGPT镜像站点

校招、秋招、春招、实习好项目！带你从零实现一个高性能的深度学习推理库，支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning inference library step by step

📖 C++11/14/17/20 Concurrency Demystified: From Core Principles to Thread-Safe Code

A fast single-producer, single-consumer lock-free queue for C++

An open-source C++ library developed and used at Facebook.

A debugging and profiling tool that can trace and visualize python code execution

Inference Llama 2 in one file of pure C

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".

LLM inference in C/C++

Inference code for Llama models

High-Resolution Image Synthesis with Latent Diffusion Models

Code for the paper "Language Models are Unsupervised Multitask Learners"

Official PyTorch implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation

Using Low-rank adaptation to quickly fine-tune diffusion models.

用 Express 和 Vue3 搭建的 ChatGPT 演示网页

FlagPerf is an open-source software platform for benchmarking AI chips.