vadim epstein

langchain-graphrag. GraphRAG / From Local to Global: A Graph RAG Approach to Query-Focused Summarization

3DCellForge. AI-powered interactive 3D model generation, inspection, and presentation studio.

MiroFish. A Simple and Universal Swarm Intelligence Engine, Predicting Anything. 简洁通用的群体智能引擎，预测万物

Emergence-World. Emergence World: A world designed to reveal what no benchmark can: emergent intelligence.

superpowers. An agentic skills framework & software development methodology that works.

BiRefNet. [CAAI AIR'24] Bilateral Reference for High-Resolution Dichotomous Image Segmentation

soul.md. The best way to build a personality for your agent. Let Claude Code / OpenClaw ingest your data & build your AI soul.

cursed_browser. True AI-Native Browser — a VLM reads the HTML and hallucinates the page.

claude-code-best-practice. from vibe coding to agentic engineering - practice makes claude perfect

relsim. 🍑 relsim: Relational Visual Similarity | pip install relsim 🌍 (CVPR 2026)

Paper2Video. Automatic Video Generation from Scientific Papers

Code2Video. [ICML 2026] Video generation via code

agenticSeek. Fully Local Manus AI. No APIs, No $200 monthly bills. Enjoy an autonomous agent that thinks, browses the web, and code for the sole cost of electricity.

HYPIR. Official implementation of HYPIR: Harnessing Diffusion-Yielded Score Priors for Image Restoration (SIGGRAPH 2025)

FlashVSR. [CVPR 2026] Towards Real-Time Diffusion-Based Streaming Video Super-Resolution — An efficient one-step diffusion framework for streaming VSR with locality-constrained sparse attention and a tiny conditional decoder.

soulx-livekit-avatar. LiveKit Talking Head Avatar with open AI S2S

Shadowbroker. Open-source intelligence for the global theater. Track everything from the corporate/private jets of the wealthy, and spy satellites, to seismic events in one unified interface. Hook an AI agent up to have it parse through data and find previously unseen correlations. The knowledge is available to all but rarely aggregated in the open, until now.

Exiv. Modular and extensible open source GenAI engine

ouroboros. Active mirror of https://github.com/razzant/ouroboros — open issues and PRs there.

infinite-ai-web. HTML

ComfyUI-Copilot. An AI-powered custom node for ComfyUI designed to enhance workflow automation and provide intelligent assistance

autogen. A programming framework for agentic AI

dify. Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.

Wan2GP. A fast AI Video Generator for the GPU Poor. Supports Wan 2.1/2.2, LTX-2, Qwen Image, Hunyuan Video, LTX Video and Flux.

mgrep. A calm, CLI-native way to semantically grep everything, like code, images, pdfs and more.

uv. An extremely fast Python package and project manager, written in Rust.

claude-context-local. Code search MCP for Claude Code. Make entire codebase the context for any coding agent. Embeddings are created and stored locally. No API cost.

smolagents. 🤗 smolagents: a barebones library for agents that think in code.

n8n. Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

workshops. One repository for some of the workshops I'm giving

dremio-local-agentic-ai-workshop. Python

graphiti. Build Real-Time Knowledge Graphs for AI Agents

phoenix. AI Observability & Evaluation

Self-Forcing. Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

LLM-Engineering-Essentials. Materials for the LLM Engineering Essentials course

servers. Model Context Protocol Servers

unsloth. Unsloth is a local UI for training and running Kimi K3, Gemma 4, Qwen3.6, DeepSeek, GLM and other models.

public-talk-slides.

btnpy. Jupyter Notebook

sliderspace. SliderSpace: Decomposing the Visual Capabilities of Diffusion Models

ComfyUI-TangoFlux. ComfyUI Custom Nodes for "TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching". This generates high-quality 44.1kHz audio up to 30 seconds using just a text prompt.

yt-dlp. A feature-rich command-line audio/video downloader

agent.exe. TypeScript

mochi. The best OSS video generation models, created by Genmo

Allegro. Allegro is a powerful text-to-video model that generates high-quality videos up to 6 seconds at 15 FPS and 720p resolution from simple text input.

Pyramid-Flow. [ICLR 2025] Pyramidal Flow Matching for Efficient Video Generative Modeling

developer. the first library to let you embed a developer agent in your own app!

E2B. Open-source, secure environment with real-world tools for enterprise-grade agents.

SyncthingWindowsSetup. Syncthing Windows Setup

PatchFusion. [CVPR 2024] An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation

ComfyUI-GGUF. GGUF Quantization support for native ComfyUI models

ai-digest. A CLI tool to aggregate your codebase into a single Markdown file for use with Claude Projects or custom ChatGPTs.

ControlNetPlus. ControlNet++: All-in-one ControlNet for image generations and editing!

aura-sr. AuraSR: GAN-based Super-Resolution for real-world

Fooocus. Focus on prompting and generating

LGM. [ECCV 2024 Oral] LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation.

ComfyUI-to-Python-Extension. A powerful tool that translates ComfyUI workflows into executable Python code.

FreeU. FreeU: Free Lunch in Diffusion U-Net (CVPR2024 Oral)

MarkovJunior. Probabilistic language based on pattern matching and constraint propagation, 153 examples

insightface. State-of-the-art 2D and 3D Face Analysis Project

PHDNet-Painterly-Image-Harmonization. [AAAI 2023] Painterly image harmonization in both spatial domain and frequency domain.

AnimateDiff. Official implementation of AnimateDiff.

CoDeF. [CVPR'24 Highlight] Official PyTorch implementation of CoDeF: Content Deformation Fields for Temporally Consistent Video Processing

DDPM_inversion. Official pytorch implementation of the paper: "An Edit Friendly DDPM Noise Space: Inversion and Manipulations". CVPR 2024.

GigaGAN. JavaScript

ConceptLab. Official Implementation for "ConceptLab: Creative Generation using Diffusion Prior Constraints"

style2paints. sketch + style = paints :art: (TOG2018/SIGGRAPH2018ASIA)

nerfstudio. A collaboration friendly studio for NeRFs

instant-ngp. Instant neural graphics primitives: lightning fast NeRF and more

ComfyUI. The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

ainodes-engine. Python

IF. Python

ImageReward. [NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences for Text-to-image Generation

Kandinsky-2. Kandinsky 2 — multilingual text2image latent diffusion model

tomesd. Speed up Stable Diffusion with this one simple trick!

Python-SpoutGL. Python wrapper for SpoutGL

Text2Video-Zero. [ICCV 2023 Oral] Text-to-Image Diffusion Models are Zero-Shot Video Generators

latentblending. Create butter-smooth transitions between prompts, powered by stable diffusion

choreography. 'I didn’t want to imitate anybody. Any movement I knew, I didn’t want to use.' – Pina Bausch

diffusers. 🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

phycv. PhyCV: The First Physics-inspired Computer Vision Library

sd-leap-booster. Fast finetuning using a booster model that puts the initial state to a local minimum

custom-diffusion. Custom Diffusion: Multi-Concept Customization of Text-to-Image Diffusion (CVPR 2023)

muse-maskgit-pytorch. Implementation of Muse: Text-to-Image Generation via Masked Generative Transformers, in Pytorch

stable-diffusion. A latent text-to-image diffusion model

voltaML. ⚡VoltaML is a lightweight library to convert and run your ML/DL deep learning models in high performance inference runtimes like TensorRT, TorchScript, ONNX and TVM.

pirounet. Conditional dance generator

Mubert-Text-to-Music. A simple notebook demonstrating prompt-based music generation via Mubert API

EvoGen-Prompt-Evolution. Jupyter Notebook

Cold-Diffusion-Models. Official implementation of Cold-Diffusion for different transformations in pytorch.

GFPGAN. GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.

InvokeAI. Invoke is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, and serves as the foundation for multiple commercial products.

stable-diffusion. A latent text-to-image diffusion model

SymphonyNet. Symphony Generation with Permutation Invariant Language Model

basic-pitch. A lightweight yet powerful audio-to-MIDI converter with pitch bend detection

eg3d. Python

eqgan-sa. [CVPR 2022] Improving GAN Equilibrium by Raising Spatial Awareness

MaGNET. Official repository for MaGNET, ICLR 2022

aesthetic-predictor. A linear estimator on top of clip to predict the aesthetic quality of pictures

CogVideo. text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

ASSET. ASSET: Autoregressive Semantic Scene Editing with Transformers at High Resolutions (SIGGRAPH 2022 - Journal Track)

wombot. Unofficial API and discord bot for wombo.art

lucid_stylegan3_datasets_models. artistic stylegan3 datasets and models we created during our ongoing stylegan3 trip.

PyTorch-Deep-Dream. Minimal PyTorch implementation of Deep Dream

insgen. [NeurIPS 2021] Data-Efficient Instance Generation from Instance Discrimination

latent-diffusion. High-Resolution Image Synthesis with Latent Diffusion Models

PyTorch-OptimalStyleTransfer. An unofficial PyTorch implementation of paper "A Closed-form Solution to Universal Style Transfer - ICCV 2019"

v-diffusion-pytorch. v objective diffusion inference code for PyTorch.

lucid-sonic-dreams. Python

fourier_feature_nets. Supplemental learning materials for "Fourier Feature Networks and Neural Volume Rendering"

ic_gan. Official repository for the paper "Instance-Conditioned GAN" by Arantxa Casanova, Marlene Careil, Jakob Verbeek, Michał Drożdżal, Adriana Romero-Soriano.

PaddleFormers. PaddleFormers is an easy-to-use library of pre-trained large language model zoo based on PaddlePaddle.

ThemeTransformer. The official implementation of Theme Transformer. A Theme-based music generation. IEEE TMM

DeceiveD. [NeurIPS 2021] Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Data

awesome-pretrained-stylegan3. A collection of pretrained models for StyleGAN3

shadertoy-render. Render a ShaderToy script directly to a video file.

MoCoGAN-HD. [ICLR 2021 Spotlight] A Good Image Generator Is What You Need for High-Resolution Video Synthesis

torchextractor. Feature extraction made simple with torchextractor

clipit. CLIP + VQGAN / PixelDraw

pixray. neural image generation

NansAreNumbers. An esoteric data type built entirely of NaNs.

articulated-animation. Code for Motion Representations for Articulated Animation paper

knowyourdata. A tool to help researchers and product teams understand datasets with the goal of improving data quality, and mitigating fairness and bias issues.

rife-ncnn-vulkan. RIFE, Real-Time Intermediate Flow Estimation for Video Frame Interpolation implemented with ncnn library

building-a-poor-mans-supercomputer. I've built a 4x V100 box for less than $5,500.

StarGANv2-VC. StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversion

MiXLab. Yet another multi-purpose Colab Notebook

style-transfer-pytorch. Neural style transfer in PyTorch.

copilot-clone. VSCode extension for code suggestion

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, ... 🧠

MelSpecVAE. Variational Autoencoder in the mel-spectrogram domain for one-shot audio synthesis

stegcloak. Hide secrets with invisible characters in plain text securely using passwords 🧙🏻‍♂️⭐

motion-models.

silero-models. Silero Models: pre-trained text-to-speech models made embarrassingly simple

Flare. Flaretic new repository, creative coding framework in c# and c++

torch-dreams. Flexible Feature visualization on PyTorch, for research and art :mag_right: :computer: :brain: :art:

gansformer. Generative Adversarial Transformers

DISTS. IQA: Deep Image Structure and Texture Similarity Metric

mediapy. This Python library makes it easy to display images and videos in a notebook.

PlotNeuralNet. Latex code for making neural networks diagrams

wit. WIT (Wikipedia-based Image Text) Dataset is a large multimodal multilingual dataset comprising 37M+ image-text sets with 11M+ unique images across 100+ languages.

tfpyth. Putting TensorFlow back in PyTorch, back in TensorFlow (differentiable TensorFlow PyTorch adapters).

lucent. Lucid library adapted for PyTorch

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

stylegan2-ada-pytorch. StyleGAN2-ADA - Official PyTorch implementation

avatars4all. Live real-time avatars from your webcam in the browser. No dedicated hardware or software installation needed. A pure Google Colab wrapper for live First-order-motion-model, aka Avatarify in the browser. And other Colabs providing an accessible interface for using FOMM, Wav2Lip and Liquid-warping-GAN with your own media and a rich GUI.

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

NeuralBeatbox_ML_Examples. Jupyter Notebook

choreo-graph. Learning latent graph representations of the dancing body with GNNs

SariGAN. [NeurIPS'20] Learning Semantic-aware Normalization for Generative Adversarial Networks

Dain-App. Source code for Dain-App

few-shot-gan. Few-shot adaptation of GANs.

tn2-wg. Tacotron2 + Waveglow Russian

Audio-transcriptor-russian-. [Russian] This script will split audio file on silence, transcript it with google recognition and save it in LJSpeech-1.1 dataset manner.

vvvv-texturefx-workshops. 🖼️ Materials for compositing workshops at NODE20

VL.RunwayML. C#

colmap. COLMAP - Structure-from-Motion and Multi-View Stereo

rewriting. Rewriting a Deep Generative Model, ECCV 2020 (oral). Interactive tool to directly edit the rules of a GAN to synthesize scenes with objects added, removed, or altered. Change StyleGANv2 to make extravagant eyebrows, or horses wearing hats.

stylegan2-surgery. StyleGAN2 fork with scripts and convenience modifications for creative media synthesis

gan-vis. Visualization of GAN training process

SketchSynthVideo-Simple. A video version of SketchSynth.

Sequencer. An algorithm that detects one-dimensional sequences in complex datasets

data-efficient-gans. [NeurIPS 2020] Differentiable Augmentation for Data-Efficient GAN Training

ScholarWithCode. A simple chrome extension to present the number of available code implementions (via Papers With Code) for articles listed on Google Scholar.

fourier-feature-networks. Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains

webpages-to-ebook. Create an EPUB from a list of URLs. Standing on the shoulders of Wget, Readability and Pandoc.

AGD. [ICML 2020] "AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks" by Yonggan Fu, Wuyang Chen, Haotao Wang, Haoran Li, Yingyan (Celine) Lin, and Zhangyang (Atlas) Wang.

Open4D. 4D visualization of dynamic events from unconstrained multi-view videos

obs-spout2-plugin. A Plugin for OBS Studio to enable Spout2 (https://github.com/leadedge/Spout2) input / output

adjustable-real-time-style-transfer. Python

Pyno. Python-based visual programming

instagram-3d-photo. A Chrome extension that adds a 3d photo effect to instagram pages.

RigNet. Code for SIGGRAPH 2020 paper "RigNet: Neural Rigging for Articulated Characters"

awesome-pretrained-stylegan. A collection of pre-trained StyleGAN models to download

awesome-pretrained-stylegan2. A collection of pre-trained StyleGAN 2 models to download

gamma-course. vvvv gamma course, based on processing-course repo

avatarify-python. Avatars for Zoom, Skype and other video-conferencing apps.

VL.Elementa. Collection of UI widgets for easy UI prototyping

3d-photo-inpainting. [CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting

GANalyze. The authors' official implementation of GANalyze, a framework for studying cognitive properties such as memorability, aesthetics, and emotional valence using genenerative models.

ganspace. Discovering Interpretable GAN Controls [NeurIPS 2020]

PIFu. This repository contains the code for the paper "PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization"

face-parsing.PyTorch. Using modified BiSeNet for face parsing in PyTorch

pytorch-optimizer. torch-optimizer -- collection of optimizers for Pytorch

BachGAN. Python

ru_transformers. Python

ruGPT2. Russian GPT2 model

Rotate-and-Render. Code for Rotate-and-Render: Unsupervised Photorealistic Face Rotation from Single-View Images (CVPR 2020)

synthesizing_human_like_sketches. Code for the WACV20 paper "Synthesizing human-like sketches from natural images using a conditional convolutional decoder"

LipGAN. This repository contains the codes for LipGAN. LipGAN was published as a part of the paper titled "Towards Automatic Face-to-Face Translation".

SinGAN. Official pytorch implementation of the paper: "SinGAN: Learning a Generative Model from a Single Natural Image"

colab-tricks. Tricks for Colab power users

3d-ken-burns. an implementation of 3D Ken Burns Effect from a Single Image using PyTorch

text-to-text-transfer-transformer. Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

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.

compress-fasttext. Tools for shrinking fastText models (in gensim format)

plink-plonk. Chrome Extension: A minimal auditory debugger for web pages

listen-to-transformer. An app to make it easier to explore and curate output from a Music Transformer

GANLatentDiscovery. The authors official implementation of Unsupervised Discovery of Interpretable Directions in the GAN Latent Space

pytorch_Realtime_Multi-Person_Pose_Estimation. Python