Qihang Yu

ponytail. Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

Kimi-K3. Open Frontier Intelligence

ao. PyTorch native quantization and sparsity for training and inference

rllab. rllab is a framework for developing and evaluating reinforcement learning algorithms, fully compatible with OpenAI Gym.

i1. Code release for "i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models"

DeepEP. DeepEP: an efficient expert-parallel communication library

FreqFlow. The official implementation of "Frequency-Aware Flow Matching for High-Quality Image Generation"

deltatok. [CVPR 2026 Highlight] A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens

BAR. [ICML 2026] code & model for arxiv paper "Autoregressive Image Generation with Masked Bit Modeling"

openpi. Python

Emu3.5. Native Multimodal Models are World Learners

Scale-RAE. Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders

VTP. [ECCV 2026] Towards Scalable Pre-training of Visual Tokenizers for Generation

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

FlowTok. PyTorch re-implementation of FlowTok: Flowing Seamlessly Across Text and Image Tokens

lightly-studio. Curate, Annotate, and Manage Your Data in LightlyStudio.

RAE. Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

datatrove. Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

dinov3. Reference PyTorch implementation and models for DINOv3

unitree_ros. C++

livox_ros_driver2. Livox device driver under Ros(Compatible with ros and ros2), support Lidar HAP and Mid-360.

Vision-Language-Vision. Python

NaVILA. [RSS'25] This repository is the implementation of "NaVILA: Legged Robot Vision-Language-Action Model for Navigation"

token-opt. Code for ICML 2025 Paper "Highly Compressed Tokenizer Can Generate Without Training"

SelftokTokenizer. Selftok: Discrete Visual Tokens of Autoregression, by Diffusion, and for Reasoning

GRAT. This repository includes the official implementation of our paper "Grouping First, Attending Smartly: Training-Free Acceleration for Diffusion Transformers"

xAR. This repository includes the official implementation of our paper "Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation"

Diffusion-wo-CFG. Official Implementation for Diffusion Models Without Classifier-free Guidance

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

GFT. Python

1.58bit.flux.

LARP. Official Pytorch implementation for LARP: Tokenizing Videos with a Learned Autoregressive Generative Prior (ICLR 2025 Oral).

Infinity. [CVPR 2025 Oral]Infinity ∞ : Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

maskbit. Implementation of the paper "MaskBit: Embedding-free Image Generation from Bit Tokens"

ViCaS. ViCaS: A Dataset for Combining Holistic and Pixel-level Video Understanding using Captions with Grounded Segmentation (CVPR'25)

CrossFlow. [CVPR2025] PyTorch-based reimplementation of CrossFlow, as proposed in 'Flowing from Words to Pixels: A Noise-Free Framework for Cross-Modality Evolution'

FlowAR. “FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching” FlowAR employs a simplest scale design and is compatible with any VAE.

genex. Generative World Explorer

adaptive-length-tokenizer. Adaptive Length Image Tokenization via Recurrent Allocation | How many tokens is an image worth ?

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

mar. PyTorch implementation of MAR+DiffLoss https://arxiv.org/abs/2406.11838

OmniTokenizer. [NeurIPS 2024]OmniTokenizer: one model and one weight for image-video joint tokenization.

Megatron-LM. Ongoing research training transformer models at scale

titok-pytorch. Implementation of TiTok, proposed by Bytedance in "An Image is Worth 32 Tokens for Reconstruction and Generation"

1d-tokenizer. This repo contains the code for 1D tokenizer and generator

DiMR. [NeurIPS 24] Alleviating Distortion in Image Generation via Multi-Resolution Diffusion Models

MambaOut. MambaOut: Do We Really Need Mamba for Vision? (CVPR 2025)

video2dataset. Easily create large video dataset from video urls

CVQ-VAE. [ICCV 2023] Online Clustered Codebook

content-debiased-fvd. [CVPR 2024] On the Content Bias in Fréchet Video Distance

coconut_cvpr2024. Jupyter Notebook

ViTamin. [CVPR 2024] Official implementation of "ViTamin: Designing Scalable Vision Models in the Vision-language Era"

VideoGPT. Jupyter Notebook

mistral-inference. Official inference library for Mistral models

open-muse. Open reproduction of MUSE for fast text2image generation.

LVM. Python

Axial-VS. This repo contains the code for our TMLR paper: A Simple Video Segmenter by Tracking Objects Along Axial Trajectories

mmc4. MultimodalC4 is a multimodal extension of c4 that interleaves millions of images with text.

qa-lora. Official PyTorch implementation of QA-LoRA

Emu. Emu Series: Generative Multimodal Models from BAAI

OmniScient-Model. This repo contains the code for our paper Towards Open-Ended Visual Recognition with Large Language Model

CLIPSelf. [ICLR2024 Spotlight] Code Release of CLIPSelf: Vision Transformer Distills Itself for Open-Vocabulary Dense Prediction

DeepSpeed. DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

3D-TransUNet. This is the official repository for the paper "3D TransUNet: Advancing Medical Image Segmentation through Vision Transformers"

llama_index. LlamaIndex is the leading document agent and OCR platform

DETA. Detection Transformers with Assignment

gpt4free. The official gpt4free repository | various collection of powerful language models | opus 4.6 gpt 5.3 kimi 2.5 deepseek v3.2 gemini 3

V3Det. Python

shikra. Python

lvis-api. Python API for LVIS Dataset

fc-clip. [NeurIPS 2023] This repo contains the code for our paper Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP

all-seeing. [ICLR 2024 & ECCV 2024] The All-Seeing Projects: Towards Panoptic Visual Recognition&Understanding and General Relation Comprehension of the Open World"

GPT4RoI. (ECCVW 2025)GPT4RoI: Instruction Tuning Large Language Model on Region-of-Interest

open_flamingo. An open-source framework for training large multimodal models.

kmax-deeplab. a PyTorch re-implementation of ECCV 2022 paper based on Detectron2: k-means mask Transformer.

LAVIS. LAVIS - A One-stop Library for Language-Vision Intelligence

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

llama. Inference code for Llama models

EDICT. Jupyter Notebook

Detic. Code release for "Detecting Twenty-thousand Classes using Image-level Supervision".

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

prompt-to-prompt. Jupyter Notebook

CrossAttentionControl. Unofficial implementation of "Prompt-to-Prompt Image Editing with Cross Attention Control" with Stable Diffusion

LaCLIP. [NeurIPS 2023] Text data, code and pre-trained models for paper "Improving CLIP Training with Language Rewrites"

SAN. Open-vocabulary Semantic Segmentation

MOSS. An open-source tool-augmented conversational language model from Fudan University

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

Segment-Everything-Everywhere-All-At-Once. [NeurIPS 2023] Official implementation of the paper "Segment Everything Everywhere All at Once"

prompt-pretraining. Official implementation for the paper "Prompt Pre-Training with Over Twenty-Thousand Classes for Open-Vocabulary Visual Recognition"

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

Painter. Painter & SegGPT Series: Vision Foundation Models from BAAI

UNINEXT. [CVPR'23] Universal Instance Perception as Object Discovery and Retrieval

CRIS.pytorch. An official PyTorch implementation of the CRIS paper

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

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.

ZegFormer. Official code for "Decoupling Zero-Shot Semantic Segmentation"

CAT-Seg. Official Implementation of "CAT-Seg🐱: Cost Aggregation for Open-Vocabulary Semantic Segmentation"

MaskCLIP. Official PyTorch implementation of "Extract Free Dense Labels from CLIP" (ECCV 22 Oral)

img2dataset. Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.

DiT. Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

ControlNet. Let us control diffusion models!

ODISE. Official PyTorch implementation of ODISE: Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models [CVPR 2023 Highlight]

GroupViT. Official PyTorch implementation of GroupViT: Semantic Segmentation Emerges from Text Supervision, CVPR 2022.

CutLER. Code release for "Cut and Learn for Unsupervised Object Detection and Instance Segmentation" and "VideoCutLER: Surprisingly Simple Unsupervised Video Instance Segmentation"

EVA. EVA Series: Visual Representation Fantasies from BAAI

open_clip. An open source implementation of CLIP.

PartImageNet. Introduction and scripts for the paper "PartImageNet: A Large, High-Quality Dataset of Parts" (Ju He, Shuo Yang, Shaokang Yang, Adam Kortylewski, Xiaoding Yuan, Jie-Neng Chen, Shuai Liu, Cheng Yang, Alan Yuille).

H-Deformable-DETR. [CVPR2023] This is an official implementation of paper "DETRs with Hybrid Matching".

parti.

autogluon. Fast and Accurate ML in 3 Lines of Code

deeplab2. DeepLab2 is a TensorFlow library for deep labeling, aiming to provide a unified and state-of-the-art TensorFlow codebase for dense pixel labeling tasks.

dxvk. Vulkan-based implementation of D3D8, 9, 10 and 11 for Linux / Wine

Swin-Transformer-TF. Tensorflow implementation of Swin Transformer model.

ModelsGenesis. [MICCAI 2019 Young Scientist Award] [MEDIA 2020 Best Paper Award] Models Genesis, one of the first "foundation" models in medical image analysis for multiple downstream tasks

L2B. This repository includes the official project of L2B, from our paper "Learning to Bootstrap for Combating Label Noise".

SAGE. No Parameters Left Behind: Sensitivity Guided Adaptive Learning Rate for Training Large Transformer Models (ICLR 2022)

convnext-tf. Python

sceneparsing. Development kit for MIT Scene Parsing Benchmark

fast_advprop. [ICLR 2022]: Fast AdvProp

LabelAssemble. [ISBI 2023] Official Implementation for Label-Assemble

ConvNeXt. Code release for ConvNeXt model

mae. PyTorch implementation of MAE https//arxiv.org/abs/2111.06377

Mask2Former. Code release for "Masked-attention Mask Transformer for Universal Image Segmentation"

TransMix. [CVPR 2022] This repository includes the official project for the paper: TransMix: Attend to Mix for Vision Transformers.

ViTs-vs-CNNs. [NeurIPS 2021]: Are Transformers More Robust Than CNNs? (Pytorch implementation & checkpoints)

K-Net. [NeurIPS2021] Code Release of K-Net: Towards Unified Image Segmentation

CaliCO. Code for ICCV2021 paper: Calibrating Concepts and Operations: Towards Symbolic Reasoning on Real Images

MaskFormer. Per-Pixel Classification is Not All You Need for Semantic Segmentation (NeurIPS 2021, spotlight)

InExtremIS. Official PyTorch implementation of InExtremIS

GG-Transformer. Code and models for the paper Glance-and-Gaze Vision Transformer

relabel_imagenet. Python

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

PanopticFCN. Fully Convolutional Networks for Panoptic Segmentation (CVPR2021 Oral)

detectron2. Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

noah-research. Noah Research

PixPro. Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learning, CVPR 2021

transformer-in-transformer. Implementation of Transformer in Transformer, pixel level attention paired with patch level attention for image classification, in Pytorch

IIC. Invariant Information Clustering for Unsupervised Image Classification and Segmentation

PVT. Official implementation of PVT series

pytorch-lightning. Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

irn. Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations, CVPR 2019 (Oral)

Unsupervised-Semantic-Segmentation. Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals. [ICCV 2021]

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

vissl. VISSL is FAIR's library of extensible, modular and scalable components for SOTA Self-Supervised Learning with images.

TransUNet. This repository includes the official project of TransUNet, presented in our paper: TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation.

pytorch-cnn-visualizations. Pytorch implementation of convolutional neural network visualization techniques

nerf. Code release for NeRF (Neural Radiance Fields)

T2T-ViT. ICCV2021, Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet

mmselfsup. OpenMMLab Self-Supervised Learning Toolbox and Benchmark

deit. Official DeiT repository

self-label. Self-labelling via simultaneous clustering and representation learning. (ICLR 2020)

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

ViT-pytorch. Pytorch reimplementation of the Vision Transformer (An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale)

vision-transformer-pytorch. Pytorch version of Vision Transformer (ViT) with pretrained models. This is part of CASL (https://casl-project.github.io/) and ASYML project.

DALLE-pytorch. Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

mmsegmentation. OpenMMLab Semantic Segmentation Toolbox and Benchmark.

SETR. [CVPR 2021 & IJCV 2024] Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers

Hearthstone-Deck-Tracker. A deck tracker and deck manager for Hearthstone on Windows

byol-pytorch. Usable Implementation of "Bootstrap Your Own Latent" self-supervised learning, from Deepmind, in Pytorch

MOS_Meticulous-Object-Segmentation.

swav. PyTorch implementation of SwAV https//arxiv.org/abs/2006.09882

MGMatting. This repository includes the official project of Mask Guided (MG) Matting, presented in our paper: Mask Guided Matting via Progressive Refinement Network

Adversarial_Metric_Attack. The code of "Adversarial Metric Attack for Person Re-identification"

CAKES. This repository contains the code for our AAAI2021 paper CAKES: Channel-wise Automatic KErnel Shrinking for Efficient 3D Networks.

ViP-DeepLab. Python

BNET. Batch Normalization with Enhanced Linear Transformation

AdelaiDet. AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.

vit-pytorch. Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

pixel-level-contrastive-learning. Implementation of Pixel-level Contrastive Learning, proposed in the paper "Propagate Yourself", in Pytorch

MODNet. A Trimap-Free Portrait Matting Solution in Real Time [AAAI 2022]

detr. End-to-End Object Detection with Transformers

closed-form-matting. Python implementation of A. Levin D. Lischinski and Y. Weiss. A Closed Form Solution to Natural Image Matting. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), June 2006, New York

GFM. [IJCV 2022] Bridging Composite and Real: Towards End-to-end Deep Image Matting

ShapeTextureDebiasedTraining. Code and models for the paper Shape-Texture Debiased Neural Network Training (ICLR 2021)

senet.pytorch. PyTorch implementation of SENet

CascadePSP. [CVPR 2020] CascadePSP: Toward Class-Agnostic and Very High-Resolution Segmentation via Global and Local Refinement

SynthCP. Offical code base for the ECCV oral paper "Synthesize then Compare: Detecting Failures and Anomalies for Semantic Segmentation"

self-supervised-3d-tasks. Python

unnas. Code for "Are labels necessary for neural architecture search"

PyContrast. PyTorch implementation of Contrastive Learning methods

CAM. Class Activation Mapping

mmagic. OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄: Generative-AI (AIGC), easy-to-use APIs, awsome model zoo, diffusion models, for text-to-image generation, image/video restoration/enhancement, etc.

CnC_Remastered_Collection. Command & Conquer: Remastered Collection

FeatureLearningRotNet. Python

RAdam. On the Variance of the Adaptive Learning Rate and Beyond

DetectoRS. DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution

PatchAttack. PatchAttack (ECCV 2020)

FBA_Matting. Official repository for the paper F, B, Alpha Matting

GCA-Matting. Official repository for Natural Image Matting via Guided Contextual Attention

Background-Matting. Background Matting: The World is Your Green Screen

ipyvolume. 3d plotting for Python in the Jupyter notebook based on IPython widgets using WebGL

indexnet_matting. (ICCV'19) Indices Matter: Learning to Index for Deep Image Matting

AutoNL. Code for AutoNL on ImageNet (CVPR2020)

mmaction. An open-source toolbox for action understanding based on PyTorch

3D-ResNets-PyTorch. 3D ResNets for Action Recognition (CVPR 2018)

pytorch-classification. Classification with PyTorch.

AtomNAS. [ICLR 2020]: 'AtomNAS: Fine-Grained End-to-End Neural Architecture Search'

temporal-segment-networks. Code & Models for Temporal Segment Networks (TSN) in ECCV 2016

rethinking-network-pruning. Rethinking the Value of Network Pruning (Pytorch) (ICLR 2019)

single-path-nas. Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours