Zhe Niu

lingbot-vision. Self-supervised learning for spatial perception

attention-gym. Helpful tools and examples for working with flex-attention

minit2i-torch. Official PyTorch re-implementation of MiniT2I.

TSR. [ICML 2026] Official code release for paper "Temporal Score Rescaling for Temperature Sampling in Diffusion and Flow Models"

frontend-slides. Create beautiful slides on the web using a coding agent's frontend skills

SenseNova-U1. SenseNova-U series: Native Unified Paradigm with NEO-unify from the First Principles

awesome-gpt-image-2. 🚀 World's largest GPT Image 2 prompt library, updated daily — 2000+ curated prompts with preview images, 16 languages. OpenAI's next-gen image model with pixel-perfect text rendering, cross-image consistency, and commercial-grade illustration. Free & open source.

Awesome-AIGC-Image-Video-Detection. A curated collection of the latest research and resources on AI-Generated Image and Video Detection.

LTX-2. Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.

caveman. 🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

reverse-SynthID. reverse engineering Gemini's SynthID detection

BaiduPCS-Go. iikira/BaiduPCS-Go原版基础上集成了分享链接/秒传链接转存功能

SoulX-Duplug. Plug-and-play streaming semantic VAD for real-time full-duplex spoken dialogue systems.

autoresearch. AI agents running research on single-GPU nanochat training automatically

Qwen-Image. Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

GatedDeltaNet. [ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Rule

OpenOCR. OpenOCR: An Open-Source Toolkit for General-OCR Research and Applications, integrates a unified training and evaluation benchmark, commercial-grade OCR and Document Parsing systems, and faithful reproductions of the core implementations from a wide range of academic papers.

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

hyper-connections. Attempt to make multiple residual streams from Bytedance's Hyper-Connections paper accessible to the public

Qwen3-TTS. Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice cloning.

Z-Image. Python

awesome-nanobanana-pro. 🚀 An awesome list of curated Nano Banana pro prompts and examples. Your go-to resource for mastering prompt engineering and exploring the creative potential of the Nano banana pro(Nano banana 2) AI image model.

TorchStream. A library for making PyTorch models streamable

sam3. The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

ovsam. [ECCV 2024] The official code of paper "Open-Vocabulary SAM".

FastVideo. A unified inference and post-training framework for accelerated video generation.

simple-lama-inpainting. Python

xxh. 🚀 Bring your favorite shell wherever you go through the ssh. Xonsh shell, fish, zsh, osquery and so on.

nano-vllm. Nano vLLM

V2SFlow. [ICASSP 2025] V2SFlow: Video-to-Speech Generation with Speech Decomposition and Rectified Flow

Awesome-Nano-Banana-images. A curated collection of fun and creative examples generated with Nano Banana & Nano Banana Pro🍌, Gemini-2.5-flash-image based model. We also release Nano-consistent-150K openly to support the community's development of image generation and unified models(click to website to see our blog)

openvino. OpenVINO™ is an open source toolkit for optimizing and deploying AI inference

faiss-gpu-wheels. Unofficial faiss wheel builder for NVIDIA GPU

faiss-wheels. Community-maintained faiss wheel builder

dinov3. Reference PyTorch implementation and models for DINOv3

HRM. Hierarchical Reasoning Model Official Release

FontDiffuser. [AAAI2024] FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning

gpt-oss. gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

timm-vit-lora. Pytorch implementation of LoRA layer for Vision Transformer from timm

supervision. We write your reusable computer vision tools. 💜

RADIO. Official repository for "AM-RADIO: Reduce All Domains Into One"

Wan2.2. Wan: Open and Advanced Large-Scale Video Generative Models

AugLy. A data augmentations library for audio, image, text, and video.

gpupixel. Real-time image filter engine based on GPU

count-anything. an empirical study on few-shot counting using segment anything (SAM)

awesome-salient-object-detection. A curated list of awesome resources for salient object detection (SOD), focusing more on multi-modal SOD, such as RGB-D SOD.

higgs-audio. Text-audio foundation model from Boson AI

python-magic. A python wrapper for libmagic

S3Tokenizer. Reverse Engineering of Supervised Semantic Speech Tokenizer (S3Tokenizer) proposed in CosyVoice

delayed-streams-modeling. Kyutai's Speech-To-Text and Text-To-Speech models based on the Delayed Streams Modeling framework.

ghost-2.0. A new one shot head swapping approach

ghost. A new one shot face swap approach for image and video domains

XPhoneBERT. XPhoneBERT: A Pre-trained Multilingual Model for Phoneme Representations for Text-to-Speech (INTERSPEECH 2023)

RobustVideoMatting. Robust Video Matting in PyTorch, TensorFlow, TensorFlow.js, ONNX, CoreML!

texture-vs-shape. Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness" (ICLR 2019 Oral)

model-vs-human. Benchmark your model on out-of-distribution datasets with carefully collected human comparison data (NeurIPS 2021 Oral)

rotation_trick. Python

MultiTalk. [NeurIPS 2025] Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation

UniRepLKNet. [CVPR 2024 & TPAMI 2025] UniRepLKNet

ml-fastvlm. This repository contains the official implementation of "FastVLM: Efficient Vision Encoding for Vision Language Models" - CVPR 2025

facefusion. Industry leading face manipulation platform

MeanFlow. PyTorch implementation of MeanFlow & iMF (one-step generative modeling).

chatterbox. SoTA open-source TTS

C-Optim. [ICLR 2026] When it comes to optimizers, it's always better to be safe than sorry

echomimic_v2. [CVPR 2025] EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation

Awesome-Talking-Head-Synthesis. 💬 An extensive collection of exceptional resources dedicated to the captivating world of talking face synthesis! ⭐ If you find this repo useful, please give it a star! 🤩

ONNX-NeuFlowV2-Optical-Flow. Python scripts performing optical flow estimation using the NeuFlowV2 model in ONNX.

codex. Lightweight coding agent that runs in your terminal

ultralytics. Ultralytics YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking

t-digest. A new data structure for accurate on-line accumulation of rank-based statistics such as quantiles and trimmed means

Open-Sora. Open-Sora: Democratizing Efficient Video Production for All

LatentSwap. Official Implementation of LatentSwap:An Efficient Latent Code Mapping Framework for Face Swapping

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

Notes. Build your personal knowledge base with TriliumNext Notes

Trilium. Build your personal knowledge base with Trilium Notes

FlowDec. An neural full-band audio codec for general audio sampled at 48 kHz with 7.5 kps or 4.5 kbps.

flash-attention. Fast and memory-efficient exact attention

xformers. Hackable and optimized Transformers building blocks, supporting a composable construction.

adjustText. A small library for automatical adjustment of text position in matplotlib plots to minimize overlaps.

fractalgen. PyTorch implementation of FractalGen https://arxiv.org/abs/2502.17437

TextBlob. Simple, Pythonic, text processing--Sentiment analysis, part-of-speech tagging, noun phrase extraction, translation, and more.

recognize-anything. Open-source and strong foundation image recognition models.

R1-V. Witness the aha moment of VLM with less than $3.

tweepy. Twitter for Python!

browser-use. 🌐 Make websites accessible for AI agents. Automate tasks online with ease.

lang-segment-anything. SAM with text prompt

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

DeepSeek-R1.

LLaSA_training. LLaSA: Scaling Train-time and Inference-time Compute for LLaMA-based Speech Synthesis

PyAV. Pythonic bindings for FFmpeg's libraries.

minimind-v. 👀「大模型」2小时从0训练65M参数的视觉多模态VLM！Train a 65M-parameter VLM from scratch in just 2h!

burn. Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

CosyVoice. Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.

CodeFormer. [NeurIPS 2022] Towards Robust Blind Face Restoration with Codebook Lookup Transformer

inswapper. One-click Face Swapper and Restoration powered by insightface 🔥

LLaVA. [NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

coconut-pytorch. Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch

awesome-controllable-speech-synthesis. This is an evolving repo for the paper "Towards Controllable Speech Synthesis in the Era of Large Language Models: A Systematic Survey".

plotext. plotting on terminal

HunyuanVideo. HunyuanVideo: A Systematic Framework For Large Video Generation Model

simclr. SimCLRv2 - Big Self-Supervised Models are Strong Semi-Supervised Learners

Grounded-Segment-Anything. Grounded SAM: Marrying Grounding DINO with Segment Anything & Stable Diffusion & Recognize Anything - Automatically Detect , Segment and Generate Anything

audio-captioning. Audio captioning - DCASE challenge 2023 task 6a

AudioSetCaps. A 6-million Audio-Caption Paired Dataset Built with a LLMs and ALMs-based Automatic Pipeline

segment-anything. The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

sam2. The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

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

LTX-Video. Official repository for LTX-Video

samurai. Official repository of "SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory"

CRATE. Code for CRATE (Coding RAte reduction TransformEr).

pytorch-grad-cam. Advanced AI Explainability for computer vision. Support for CNNs, Vision Transformers, Classification, Object detection, Segmentation, Image similarity and more.

Cosmos-Tokenizer. A suite of image and video neural tokenizers

VideoTuna. Let's finetune video generation models!

bypy. Python client for Baidu Yun (Personal Cloud Storage) 百度云/百度网盘Python客户端

InsightFace-REST. InsightFace REST API for easy deployment of face recognition services with TensorRT in Docker.

torchcodec. PyTorch media decoding and encoding

dacite. Simple creation of data classes from dictionaries.

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

libfacedetection. An open source library for face detection in images. The face detection speed can reach 1000FPS.

pintora. An extensible text-to-diagrams library that works in both browser and node.js

face_detection. Python

retinaface. RetinaFace: Deep Face Detection Library for Python

Analytic-DPM. Code for the paper Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models (ICLR 2022 Outstanding Paper Award)

ddpo. Code for the paper "Training Diffusion Models with Reinforcement Learning"

minGRU-pytorch. Implementation of the proposed minGRU in Pytorch

open_clip. An open source implementation of CLIP.

dpm-solver. Official code for "DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps" (Neurips 2022 Oral)

termplotlib. :chart_with_upwards_trend: Plotting on the command line

flux. Official inference repo for FLUX.1 models

sidbench. A Python framework for reliably assessing synthetic image detection methods

DeepFakeDefenders. Image forgery recognition algorithm

awesome-talking-head-generation.

mini-omni. open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.

playwright-python. Python version of the Playwright testing and automation library.

mi-gpt. 🏠 将小爱音箱接入 ChatGPT 和豆包，改造成你的专属语音助手。

transfusion-pytorch. Pytorch implementation of Transfusion, "Predict the Next Token and Diffuse Images with One Multi-Modal Model", from MetaAI

LossRider. A plotting tool that outputs Line Rider maps, so you can watch a man on a sled scoot down your loss curves. 🎿

unsloth. Unsloth Studio is a web UI for training and running open models like Gemma 4, Qwen3.6, DeepSeek, gpt-oss locally.

llama3. The official Meta Llama 3 GitHub site

LaTeXML. LaTeXML: a TeX and LaTeX to XML/HTML/ePub/MathML translator.

ema-pytorch. A simple way to keep track of an Exponential Moving Average (EMA) version of your Pytorch model

edm2. EDM2 and Autoguidance -- Official PyTorch implementation

Languagecodec. [ACL 2025 Oral] Language-Codec: Reducing the Gaps Between Discrete Codec Representation and Speech Language Models

descript-audio-vae. VAE modified from Descript Audio Codec, which replaces the RVQ with VAE

VAR. [NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!

tango. A family of diffusion models for text-to-audio generation.

AudioLCM. PyTorch Implementation of AudioLCM (ACM-MM'24): a efficient and high-quality text-to-audio generation with latent consistency model.

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

infomap. Multi-level network clustering based on the Map Equation

SimPO. [NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward

vocos. Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis

ChatTTS. A generative speech model for daily dialogue.

LoRA. Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

PowerInfer. High-speed Large Language Model Serving for Local Deployment

mamba. Mamba SSM architecture

OpenVoice. Instant voice cloning by MIT and MyShell. Audio foundation model.

AnyText. Official implementation code of the paper <AnyText: Multilingual Visual Text Generation And Editing>

normalizing-flows. PyTorch implementation of normalizing flow models

ml-sigma-reparam. Python

diffusion_schrodinger_bridge. PyTorch Implementation of DSB for Score Based Generative Modeling. Experiments managed using Hydra.

resemble-enhance. AI powered speech denoising and enhancement

nanoRWKV. RWKV in nanoGPT style

gdrive. Google Drive CLI Client

Qwen-Audio. The official repo of Qwen-Audio (通义千问-Audio) chat & pretrained large audio language model proposed by Alibaba Cloud.

vall-e. An unofficial PyTorch implementation of VALL-E

llm.nvim. LLM powered development for Neovim

voicebox-pytorch. Implementation of Voicebox, new SOTA Text-to-speech network from MetaAI, in Pytorch

AudioLDM2. Text-to-Audio/Music Generation

VALL-E-X. An open source implementation of Microsoft's VALL-E X zero-shot TTS model. Demo is available in https://plachtaa.github.io/vallex/

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

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

SC_VALL-E. Style-Controllable Zero-Shot Text to Speech Synthesizer based on VALL-E

conditional-flow-matching. TorchCFM: a Conditional Flow Matching library

aim. Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.

SmallInitEmb. LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence

linkgan.

Pengi. An Audio Language model for Audio Tasks

SpeechGPT. SpeechGPT Series: Speech Large Language Models

AcademiCodec. AcademiCodec: An Open Source Audio Codec Model for Academic Research

wttr.in. :partly_sunny: The right way to check the weather

I2SB. Python

InstructTTS. The deme page of InstructTTS

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

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

AutoGPT. AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

bark. 🔊 Text-Prompted Generative Audio Model

consistency_models. Official repo for consistency models.

alpaca-lora. Instruct-tune LLaMA on consumer hardware

typst. A markup-based typesetting system that is powerful and easy to learn.

LMFlow. An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.

torch-audiomentations. Fast audio data augmentation in PyTorch. Inspired by audiomentations. Useful for deep learning.

gpt4all. GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.

RWKV-LM. RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

Watson. :watch: A wonderful CLI to track your time!

attention_with_linear_biases. Code for the ALiBi method for transformer language models (ICLR 2022)

nimtorch. PyTorch - Python + Nim

llama. Inference code for Llama models

speech-synthesis-paper. List of speech synthesis papers.

voxceleb_trainer. In defence of metric learning for speaker recognition