Wensen Wu

ghostbuster. Ghostbuster: Detecting Text Ghostwritten by Large Language Models (NAACL 2024)

ToolBench. [ICLR'24 spotlight] An open platform for training, serving, and evaluating large language model for tool learning.

Awesome-LLM-Inference. 📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention, WINT8/4, Parallelism, etc.🎉

llm-analysis. Latency and Memory Analysis of Transformer Models for Training and Inference

ARES. Automated Evaluation of RAG Systems

llmperf. LLMPerf is a library for validating and benchmarking LLMs

awesome-ai-agents. A list of AI autonomous agents

RLPHF. Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging

Awesome-LLM-Compression. Awesome LLM compression research papers and tools.

gisting. Learning to Compress Prompts with Gist Tokens - https://arxiv.org/abs/2304.08467

TransformerEngine. A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.

ml-engineering. Machine Learning Engineering Open Book

manifest. Prompt programming with FMs.

lilac. Curate better data for LLMs

RedPajama-Data. The RedPajama-Data repository contains code for preparing large datasets for training large language models.

vimGPT. Browse the web with GPT-4V and Vimium

dino. PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

TensorRT-LLM. TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.

Wonder3D. Single Image to 3D using Cross-Domain Diffusion for 3D Generation

llm-awq. [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

umap. Uniform Manifold Approximation and Projection

video-retalking. [SIGGRAPH Asia 2022] VideoReTalking: Audio-based Lip Synchronization for Talking Head Video Editing In the Wild

openchat. OpenChat: Advancing Open-source Language Models with Imperfect Data

skypilot. Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

triton. Development repository for the Triton language and compiler

tesseract. Tesseract Open Source OCR Engine (main repository)

llama2.c. Inference Llama 2 in one file of pure C

stable-diffusion.cpp. Diffusion model(SD,Flux,Wan,Qwen Image,Z-Image,...) inference in pure C/C++

llavavision. A simple "Be My Eyes" web app with a llama.cpp/llava backend

animatediff-cli. a CLI utility/library for AnimateDiff stable diffusion generation

realtime-bakllava. llama.cpp with BakLLaVA model describes what does it see

CogVLM. a state-of-the-art-level open visual language model | 多模态预训练模型

AutoAWQ. AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:

localpilot. Python

ggml. Tensor library for machine learning

dinov2. PyTorch code and models for the DINOv2 self-supervised learning method.

axolotl. Go ahead and axolotl questions

Awesome-Efficient-LLM. A curated list for Efficient Large Language Models

Awesome-LLM. Awesome-LLM: a curated list of Large Language Model

gptq. Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".

FLAML. A fast library for AutoML and tuning. Join our Discord: https://discord.gg/Cppx2vSPVP.

autogen. A programming framework for agentic AI

AgentTuning. AgentTuning: Enabling Generalized Agent Abilities for LLMs

S-LoRA. S-LoRA: Serving Thousands of Concurrent LoRA Adapters

AutoGPTQ. An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

LLM-Shearing. [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning

consistencydecoder. Consistency Distilled Diff VAE

ollama. Get up and running with Kimi-K2.6, GLM-5.1, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

web-stable-diffusion. Bringing stable diffusion models to web browsers. Everything runs inside the browser with no server support.

Yi. A series of large language models trained from scratch by developers @01-ai

sshx. Fast, collaborative live terminal sharing over the web

stable-fast. https://wavespeed.ai/ Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.

openinterpreter. A lightweight coding agent for open models like Deepseek, Kimi, and Qwen

streaming-llm. [ICLR 2024] Efficient Streaming Language Models with Attention Sinks

MetaGPT. 🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

evals. Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.

langchain. The agent engineering platform.

natbot. Drive a browser with GPT-3

whisper. Robust Speech Recognition via Large-Scale Weak Supervision

torchsparse. [MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

artbench. Benchmarking Generative Models with Artworks

dalle-mini. DALL·E Mini - Generate images from a text prompt

makemore. An autoregressive character-level language model for making more things

Cytoid. A community-driven touchscreen music game.

RetroArch. Cross-platform, sophisticated frontend for the libretro API. Licensed GPLv3.

rich. Rich is a Python library for rich text and beautiful formatting in the terminal.

github1s. One second to read GitHub code with VS Code.

SingleFile. Web Extension for saving a faithful copy of a complete web page in a single HTML file

evolution. A sandbox game/simulator that demonstrates machine learning with evolutionary algorithms.

jspaint. 🎨 Classic MS Paint, ＲＥＶＩＶＥＤ + ✨Extras

cool-retro-term. A good looking terminal emulator which mimics the old cathode display...

pyxel. A retro game engine for Python

bevy. A refreshingly simple data-driven game engine built in Rust

godot. Godot Engine – Multi-platform 2D and 3D game engine

TIC-80. TIC-80 is a fantasy computer for making, playing and sharing tiny games.

isowords. Open source game built in SwiftUI and the Composable Architecture.

gameboy.live. 🕹️ A basic gameboy emulator with terminal "Cloud Gaming" support

ldtk. Modern, lightweight and efficient 2D level editor

diffuser. Code for the paper "Planning with Diffusion for Flexible Behavior Synthesis"

Simple-Gallery. A premium app for managing and editing your photos, videos, GIFs without ads

solar. Pelican theme for solar powered version of lowtechmagazine.com

Celeste. Celeste Bugs & Issue Tracker + some Source Code

annotated_deep_learning_paper_implementations. 🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

neox. Simple Annotated implementation of GPT-NeoX in PyTorch

gpt-neox. An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

WebGL-Fluid-Simulation. Play with fluids in your browser (works even on mobile)

eigen. The Art World in Your Pocket or Your Trendy Tech Company's Tote, Artsy's mobile app.

shan-shui-inf. Procedurally generated Chinese landscape painting.

awesome-design. 🌟 Curated design resources from all over the world.

awesome-creative-coding. Creative Coding: Generative Art, Data visualization, Interaction Design, Resources.

open-source-ios-apps. :iphone: Collaborative List of Open-Source iOS Apps

MarbleMarcher. A Fractal Physics Game

bitsy. a little engine for little games, worlds, and stories

dotfiles. Dot files for Evan Chen (Arch Linux on i3)

seaport. Seaport is a marketplace protocol for safely and efficiently buying and selling NFTs.

rust-autograd. Tensors and differentiable operations (like TensorFlow) in Rust

memorizing-transformers-pytorch. Implementation of Memorizing Transformers (ICLR 2022), attention net augmented with indexing and retrieval of memories using approximate nearest neighbors, in Pytorch

nvdiffrec. Official code for the CVPR 2022 (oral) paper "Extracting Triangular 3D Models, Materials, and Lighting From Images".

rayon. Rayon: A data parallelism library for Rust

tauri. Build smaller, faster, and more secure desktop and mobile applications with a web frontend.

OS.js. OS.js - JavaScript Web Desktop Platform

pce. Emulates Mac Plus, PC, & Atari ST in the browser using WebAssembly

serenity. The Serenity Operating System 🐞

v86. x86 PC emulator and x86-to-wasm JIT, running in the browser

anomalib. An anomaly detection library comprising state-of-the-art algorithms and features such as experiment management, hyper-parameter optimization, and edge inference.

ml-system-design-pattern. System design patterns for machine learning

code-server. VS Code in the browser

Awesome-LLMOps. An awesome & curated list of best LLMOps tools for developers

mixture-of-experts. PyTorch Re-Implementation of "The Sparsely-Gated Mixture-of-Experts Layer" by Noam Shazeer et al. https://arxiv.org/abs/1701.06538

ColossalAI. Making large AI models cheaper, faster and more accessible

next-mdx-blog. Next.js + MDX blog template with Tailwind CSS and TypeScript.

meetsy. Open source online booking system

SymphonyNet. Symphony Generation with Permutation Invariant Language Model

folio-2019. JavaScript

vditor. ♏ 一款浏览器端的 Markdown 编辑器，支持所见即所得（富文本）、即时渲染（类似 Typora）和分屏预览模式。An In-browser Markdown editor, support WYSIWYG (Rich Text), Instant Rendering (Typora-like) and Split View modes.

AppFlowy. Bring projects, wikis, and teams together with AI. AppFlowy is the AI collaborative workspace where you achieve more without losing control of your data. The leading open source Notion alternative.

inspire. Collection of frontend dev and web design links :bulb:

turborepo. Build system optimized for JavaScript and TypeScript, written in Rust

unilm. Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

supabase. The Postgres development platform. Supabase gives you a dedicated Postgres database to build your web, mobile, and AI applications.

juice. The Hacker's Machine Learning Engine

daedalOS. Desktop environment in the browser

ProjectVisBug. FireBug for designers › Edit any webpage, in any state https://a.nerdy.dev/gimme-visbug

squoosh. Make images smaller using best-in-class codecs, right in the browser.

darts. A python library for user-friendly forecasting and anomaly detection on time series.

faiss. A library for efficient similarity search and clustering of dense vectors.

google-research. Google Research

pyod. A Python library for anomaly detection across tabular, time series, graph, text, image, and audio data. 60+ detectors, benchmark-backed ADEngine orchestration, and an agentic workflow for AI agents.

bubbletea. A powerful little TUI framework 🏗

metrics. 📊 An infographics generator with 30+ plugins and 300+ options to display stats about your GitHub account and render them as SVG, Markdown, PDF or JSON!

Awesome-Diffusion-Models. A collection of resources and papers on Diffusion Models

awesome-readme. A curated list of awesome READMEs

MiCE. Pytorch implementation for ICLR 2021 paper - MiCE: Mixture of Contrastive Experts for Unsupervised Image Clustering

optimate. A collection of libraries to optimise AI model performances

strml.net. STRML: Projects & Work

tabby. A terminal for a more modern age

patchcore-inspection. Python

StyleCLIP. Official Implementation for "StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery" (ICCV 2021 Oral)

Telegram. Telegram for Android source

conky. Light-weight system monitor for X, Wayland, and other things, too

DALLE2-pytorch. Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network, in Pytorch

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

zsh-iterm-touchbar. Display feedback of terminal in the 🍏 Touchbar

neuronika. Tensors and dynamic neural networks in pure Rust.

enso. Enso Analytics is a self-service data prep and analysis platform designed for data teams.

tch-rs. Rust bindings for the C++ api of PyTorch.

hyper. A terminal built on web technologies

neovim. Vim-fork focused on extensibility and usability

kitty. If you live in the terminal, kitty is made for you! Cross-platform, fast, feature-rich, GPU based.

fd. A simple, fast and user-friendly alternative to 'find'

ripgrep. ripgrep recursively searches directories for a regex pattern while respecting your gitignore

zoxide. A smarter cd command. Supports all major shells.

kubernetes. Production-Grade Container Scheduling and Management

quick-look-plugins. List of useful Quick Look plugins for developers

the_silver_searcher. A code-searching tool similar to ack, but faster.

ml4code.github.io. Website for "A Survey of Machine Learning for Big Code and Naturalness"

data. Data and code behind the articles and graphics at FiveThirtyEight

awesome-computer-vision. A curated list of awesome computer vision resources

go-web-framework-stars. :star: Web frameworks for Go, most starred on GitHub

vector. A high-performance observability data pipeline.

awesome-tunneling. List of ngrok, Cloudflare Tunnel, Tailscale, and ZeroTier alternatives and other tunneling software and services. Focus on self-hosting.

hsd. Handshake Daemon & Full Node

awesome. 😎 Awesome lists about all kinds of interesting topics

ALL-about-RSS. A list of RSS related stuff: tools, services, communities and tutorials, etc.

RIDE-Swin-LongTailRecognition. Swin-Transformer-based version of RIDE (ICLR 2021 Spotlight)

metaseq. Repo for external large-scale work

monaco-editor. A browser based code editor

PlacesToPostYourStartup. Compiled list of links from "Ask HN: Where can I post my startup to get beta users?"

clip-as-service. 🏄 Scalable embedding, reasoning, ranking for images and sentences with CLIP

ncnn. ncnn is a high-performance neural network inference framework optimized for the mobile platform

dlrm. An implementation of a deep learning recommendation model (DLRM)

disco-diffusion. Jupyter Notebook

dalle-2-preview.

Awesome-Incremental-Learning. Awesome Incremental Learning

Megatron-LM. Ongoing research training transformer models at scale

gpt-2. Code for the paper "Language Models are Unsupervised Multitask Learners"

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

libzmq. ZeroMQ core engine in C++, implements ZMTP/3.1