Open-source deep-learning software
2785 open-source libraries, tools and applications tagged deep-learning, ranked by how closely deep-learning is their primary purpose, then by adoption and maintenance health. Each entry links to a page with sources, use cases and when to choose or avoid it.
2785 products, primary matches first, then adoption-weighted; health v2 shown.
| Product | Health v2 | Stars | Maturity |
|---|---|---|---|
| pytorch/torchdynamo TorchDynamo is a Python-level JIT compiler that speeds up unmodified PyTorch programs by capturing Python bytecode into FX graphs. The proj… | 10 | 1078 | maintenance |
| floodsung/LearningToCompare_FSL Official PyTorch implementation of the Relation Network for Few-Shot Learning from the CVPR 2018 paper 'Learning to Compare'. It includes t… | 32 | 1076 | maintenance |
| DirectVoxGO DirectVoxGO (DVGO) is a PyTorch implementation of Direct Voxel Grid Optimization for fast neural radiance field (NeRF) reconstruction, repl… | 32 | 1075 | maintenance |
| JosephKJ/OWOD OWOD is the official PyTorch implementation of ORE (Open World Object Detector), a CVPR 2021 Oral paper introducing the Open World Object D… | 32 | 1073 | maintenance |
| YudongGuo/AD-NeRF A PyTorch implementation of AD-NeRF, an ICCV 2021 paper that synthesizes talking-head videos by driving neural radiance fields with audio i… | 32 | 1073 | maintenance |
| cardwing/Codes-for-Lane-Detection Reference implementations of lightweight lane detection CNNs, including the ENet-SAD model from the ICCV 2019 paper 'Learning Lightweight L… | 32 | 1072 | maintenance |
| megvii-research/PETR Official PyTorch implementation of PETR and PETRv2, transformer-based frameworks for multi-view 3D object detection and BEV segmentation fr… | 32 | 1071 | maintenance |
| gusye1234/LightGCN-PyTorch A PyTorch implementation of LightGCN, a simplified graph convolution network for collaborative filtering recommendation from the SIGIR 2020… | 10 | 1071 | maintenance |
| SergeyMorugin/ostagram Ostagram is a self-hosted Ruby on Rails web application that applies neural style transfer to user-uploaded photos, blending them with the … | 47 | 1068 | maintenance |
| MatthieuCourbariaux/BinaryNet BinaryNet is a research codebase for training deep neural networks whose weights and activations are constrained to +1 or -1, reproducing t… | 32 | 1068 | maintenance |
| omerbt/MultiDiffusion Official PyTorch implementation of MultiDiffusion (ICML 2023), a training-free framework that fuses multiple diffusion paths over a pre-tra… | 32 | 1068 | maintenance |
| NVIDIA-AI-IOT/trt_pose trt_pose is a Python library from NVIDIA for real-time human pose estimation accelerated with TensorRT, targeting NVIDIA Jetson and other N… | 23 | 1067 | maintenance |
| openai/blocksparse A Python package of TensorFlow ops and custom CUDA GPU kernels for block-sparse matrix multiplication and convolution, plus related ops lik… | 10 | 1067 | maintenance |
| deepmedic/deepmedic DeepMedic is an efficient multi-scale 3D convolutional neural network for segmenting 3D medical scans such as MRI and CT. It is a Python-ba… | 23 | 1066 | maintenance |
| ternaus/TernausNet TernausNet is a PyTorch implementation of the U-Net architecture with a VGG11 encoder pre-trained on ImageNet for image segmentation. It wa… | 32 | 1062 | maintenance |
| open-mmlab/mmflow MMFlow is an open-source optical flow toolbox and benchmark built on PyTorch as part of the OpenMMLab project. It provides a unified, modul… | 23 | 1061 | maintenance |
| RexYing/gnn-model-explainer Official implementation of GNNExplainer, a NeurIPS 2019 method for explaining predictions of graph neural networks by learning soft masks o… | 32 | 1060 | maintenance |
| OFA-Sys/ONE-PEACE ONE-PEACE is a general multimodal representation model that jointly encodes vision, audio, and language modalities without initializing fro… | 30 | 1060 | maintenance |
| philipperemy/keract Keract is a Python library that extracts layer activations (outputs) and gradients from TensorFlow/Keras models as NumPy arrays, with helpe… | 26 | 1060 | maintenance |
| zhaoweicai/cascade-rcnn A C++/Caffe implementation of Cascade R-CNN and other popular two-stage object detection frameworks such as Faster R-CNN, R-FCN, and FPN. I… | 32 | 1059 | maintenance |
| JiehangXie/PaddleBoBo PaddleBoBo is a Python project built on PaddlePaddle (with PaddleSpeech and PaddleGAN) that quickly generates a virtual streamer (VTuber) f… | 32 | 1057 | maintenance |
| chenyuntc/PyTorchText A PyTorch implementation of multiple text classification models (TextCNN, TextRNN/LSTM, RCNN, FastText, inception CNN) that won 1st place i… | 32 | 1057 | maintenance |
| piergiaj/pytorch-i3d A PyTorch port of DeepMind's I3D (Inflated 3D ConvNet) models pretrained on the Kinetics dataset for video action recognition. It includes … | 32 | 1057 | maintenance |
| HRNet/HRNet-Image-Classification Official PyTorch implementation and training code for HRNet (High-Resolution Network) image classification models on ImageNet. It provides … | 23 | 1057 | maintenance |
| damo-cv/TransReID Official PyTorch implementation of TransReID, an ICCV 2021 paper applying vision transformers to object re-identification. It provides trai… | 32 | 1056 | maintenance |
| raminmh/CfC Reference implementations of Closed-form Continuous-time (CfC) neural networks, a fast closed-form approximation of liquid time-constant ne… | 23 | 1056 | maintenance |
| keras-team/keras-cv KerasCV is a library of modular computer vision components built on Keras 3 that work natively with TensorFlow, JAX, or PyTorch. It provide… | 10 | 1055 | maintenance |
| microsoft/Oscar Oscar is Microsoft's research code for object-semantics aligned cross-modal pre-training of vision-language models, with VinVL providing im… | 10 | 1054 | maintenance |
| huggingface/optimum-quanto Optimum Quanto is a PyTorch quantization backend for Hugging Face Optimum that quantizes model weights (int2/int4/int8/float8) and activati… | 66 | 1053 | maintenance |
| 4uiiurz1/pytorch-nested-unet A PyTorch implementation of the UNet++ (Nested U-Net) architecture for image segmentation, based on the paper 'UNet++: A Nested U-Net Archi… | 32 | 1051 | maintenance |
| thunlp/OpenDelta OpenDelta is a Python library for parameter-efficient tuning (delta tuning) of pretrained language models, letting users attach small train… | 23 | 1048 | maintenance |
| microsoft/SimMIM Official PyTorch implementation of SimMIM, a simple framework for masked image modeling (self-supervised visual pre-training) from Microsof… | 10 | 1048 | maintenance |
| yangxudong/deeplearning A collection of deep learning model training, evaluation, and prediction code implemented with TensorFlow's high-level Estimator API, with … | 32 | 1047 | maintenance |
| YuwenXiong/py-R-FCN A Python implementation of R-FCN (Region-based Fully Convolutional Networks) for object detection, modified from the official MATLAB code a… | 32 | 1044 | maintenance |
| JIA-Lab-research/SNR-Aware-Low-Light-Enhance Official PyTorch implementation of the CVPR 2022 paper 'SNR-aware Low-Light Image Enhancement'. It combines SNR-aware transformers and conv… | 32 | 1044 | maintenance |
| CalciferZh/minimal-hand A Python library for real-time 3D hand motion capture from a single RGB camera, running at over 100fps. It combines a deep network (DetNet)… | 23 | 1043 | maintenance |
| facebookresearch/FixRes FixRes is a PyTorch implementation of the NeurIPS 2019 paper 'Fixing the train-test resolution discrepancy', providing training and fine-tu… | 10 | 1043 | maintenance |
| google-research/morph-net MorphNet is a TensorFlow library for resource-constrained learning of deep network structure. It adds regularizers during training that ind… | 76 | 1042 | maintenance |
| asingh33/CNNGestureRecognizer A desktop application that recognizes hand gestures from webcam video using a convolutional neural network built with Keras, TensorFlow/The… | 57 | 1042 | maintenance |
| SysCV/sam-pt SAM-PT extends the Segment Anything Model to zero-shot video segmentation by combining SAM with sparse point-based tracking (PIPS, CoTracke… | 29 | 1042 | maintenance |
| descriptinc/melgan-neurips Official PyTorch implementation of MelGAN, a GAN-based non-autoregressive vocoder that inverts mel-spectrograms into raw audio waveforms fo… | 32 | 1041 | maintenance |
| huggingface/pytorch-pretrained-BigGAN A PyTorch reimplementation of DeepMind's BigGAN generator with pretrained weights at 128, 256, and 512 pixel resolutions, plus scripts to c… | 10 | 1041 | maintenance |
| google-research/deeplab2 DeepLab2 is a TensorFlow library from Google Research providing a unified, state-of-the-art codebase for dense pixel labeling tasks such as… | 10 | 1038 | maintenance |
| Kismuz/btgym BTGym is a Python library that wraps the Backtrader algorithmic trading engine in an OpenAI Gym environment API, enabling event-driven back… | 32 | 1037 | maintenance |
| keijiro/Pix2Pix A Unity library that runs pix2pix image-to-image translation neural networks in real time using compute shaders. It includes its own infere… | 23 | 1037 | maintenance |
| sicara/tf-explain tf-explain is a Python library implementing interpretability methods (Grad CAM, occlusion sensitivity, vanilla gradients, etc.) for tf.kera… | 23 | 1035 | maintenance |
| google-deepmind/scalable_agent A TensorFlow implementation of IMPALA (Importance Weighted Actor-Learner Architectures) for scalable distributed deep reinforcement learnin… | 32 | 1034 | maintenance |
| pesser/stable-diffusion The development repository for Stable Diffusion and Latent Diffusion Models, containing research code, training scripts, and pretrained mod… | 32 | 1032 | maintenance |
| goberoi/faceit A Python script that simplifies swapping faces in videos using the deepfakes/faceswap library, with training data sourced from YouTube vide… | 32 | 1032 | maintenance |
| ClementPinard/SfmLearner-Pytorch A PyTorch reimplementation of SfMLearner (Zhou et al., CVPR 2017), which jointly learns monocular depth estimation and camera ego-motion fr… | 63 | 1031 | maintenance |
| qubvel/ttach TTAch is a Python library for image test time augmentation (TTA) with PyTorch. It wraps existing models to apply augmentations like flips, … | 23 | 1031 | maintenance |
| EdgeCrafter DEIMv2 is a real-time object detection framework that extends the DEIM DETR family with DINOv3-pretrained and distilled backbones plus a Sp… | 59 | 364 | active |
| xingyizhou/ExtremeNet Official PyTorch implementation of ExtremeNet, a CVPR 2019 bottom-up object detection method that detects four extreme points and one cente… | 32 | 1030 | maintenance |
| graphdeeplearning/graphtransformer Reference implementation of the Graph Transformer architecture from the paper 'A Generalization of Transformer Networks to Graphs' (DLG-AAA… | 32 | 1030 | maintenance |
| vincentherrmann/pytorch-wavenet A PyTorch implementation of the WaveNet neural audio generation architecture with fast generation support. It includes dataset creation fro… | 32 | 1029 | maintenance |
| yuval-alaluf/hyperstyle Official PyTorch implementation of HyperStyle (CVPR 2022), a hypernetwork that inverts real images into editable regions of StyleGAN's late… | 32 | 1029 | maintenance |
| kakaobrain/rq-vae-transformer The official PyTorch implementation of 'Autoregressive Image Generation using Residual Quantization' (CVPR 2022), implementing RQ-VAE and R… | 32 | 1028 | maintenance |
| lmb-freiburg/flownet2 A Caffe fork implementing FlowNet 2.0, a deep CNN for optical flow estimation from image pairs, released with the CVPR 2017 paper. It inclu… | 32 | 1026 | maintenance |
| turtlesoupy/this-word-does-not-exist A project that trains a GPT-2 variant to invent fake English words with generated definitions and example sentences, powering the thiswordd… | 68 | 1024 | maintenance |
| awslabs/multi-model-server Multi Model Server (MMS) is a tool for serving deep learning model inference over HTTP endpoints, supporting models from any ML/DL framewor… | 10 | 1024 | maintenance |
| sniklaus/sepconv-slomo A reference PyTorch implementation of Video Frame Interpolation via Adaptive Separable Convolution, which generates intermediate frames bet… | 39 | 1021 | maintenance |
| nflows nflows is a comprehensive PyTorch library of normalizing flows for density estimation and generative modeling. It provides invertible trans… | 23 | 1021 | maintenance |
| Gumpest/YOLOv5-Multibackbone-Compression A YOLOv5-based toolbox for swapping in lightweight or high-accuracy backbones (TPH-YOLOv5, GhostNet, ShuffleNetV2, MobileNetV3-Small, Effic… | 32 | 1020 | maintenance |
| EvgenyKashin/stylegan2-distillation A research implementation of the ECCV 2020 paper 'StyleGAN2 Distillation for Feed-forward Image Manipulation', distilling StyleGAN2 latent-… | 32 | 1018 | maintenance |
| tensorflow/similarity TensorFlow Similarity is a Python library built on TensorFlow for training and serving similarity and metric learning models, including sel… | 10 | 1018 | maintenance |
| brain-research/self-attention-gan A TensorFlow implementation of Self-Attention GANs for reproducing results from the paper 'Self-Attention Generative Adversarial Networks' … | 10 | 1018 | maintenance |
| facebookresearch/barlowtwins A PyTorch implementation of Barlow Twins, a self-supervised learning method that trains visual representations by reducing redundancy betwe… | 10 | 1015 | maintenance |
| AMLab-Amsterdam/AttentionDeepMIL A PyTorch implementation of the paper 'Attention-based Deep Multiple Instance Learning' (Ilse, Tomczak & Welling, 2018). It provides code t… | 32 | 1013 | maintenance |
| Oldpan/Pytorch-Memory-Utils A small set of Python scripts for tracking GPU memory usage in PyTorch, including per-line tensor allocation reports and model parameter/in… | 32 | 1013 | maintenance |
| wpeebles/gangealing Official PyTorch implementation of GANgealing, a CVPR 2022 method that trains a Spatial Transformer to densely align images using GAN-gener… | 32 | 1013 | maintenance |
| snap-research/NeROIC Official PyTorch implementation of NeROIC, a neural method for capturing 3D object geometry and material from online image collections and … | 32 | 1012 | maintenance |
| tensorflow/neural-structured-learning Neural Structured Learning (NSL) is a TensorFlow framework for training neural networks with structured signals, either explicit graphs or … | 61 | 1011 | maintenance |
| allegro/allRank allRank is a PyTorch-based framework for training neural learning-to-rank (LTR) models. It provides pointwise, pairwise, and listwise loss … | 32 | 1011 | maintenance |
| bubbliiiing/yolov8-pytorch A PyTorch implementation of the YOLOv8 object detection model with training, prediction, and evaluation scripts. It supports training on cu… | 22 | 1011 | maintenance |
| xiaoyufenfei/Efficient-Segmentation-Networks A PyTorch reference implementation collection of lightweight, real-time semantic segmentation models such as ENet, ERFNet, LEDNet, Fast-SCN… | 32 | 1009 | maintenance |
| zhanghang1989/PyTorch-Multi-Style-Transfer A PyTorch implementation of MSG-Net and Gatys et al. neural style transfer for applying artistic styles to images in real time. It includes… | 23 | 1009 | maintenance |
| mileyan/pseudo_lidar Research code implementing Pseudo-LiDAR, a CVPR 2019 method that converts image-based depth maps into pseudo-LiDAR point clouds for 3D obje… | 32 | 1007 | maintenance |
| PeterWang512/CNNDetection A PyTorch research codebase with pretrained models for detecting CNN-generated (GAN/synthetic) images, from the CVPR 2020 paper 'CNN-genera… | 32 | 1006 | maintenance |
| alex04072000/ObstructionRemoval The official TensorFlow implementation of the CVPR 2020 paper 'Learning to See Through Obstructions', which removes obstructions like windo… | 32 | 1005 | maintenance |
| microsoft/nnfusion NNFusion is a flexible and efficient deep neural network (DNN) compiler that generates high-performance executables from model descriptions… | 10 | 1005 | maintenance |
| kevinzakka/spatial-transformer-network A TensorFlow implementation of Spatial Transformer Networks, a differentiable module that can be inserted into ConvNet architectures to add… | 32 | 1004 | maintenance |
| johannakarras/DreamPose Official PyTorch implementation of DreamPose, a Stable Diffusion-based model that synthesizes animated fashion videos from a single image a… | 31 | 1004 | maintenance |
| NATSpeech/NATSpeech A PyTorch framework for non-autoregressive text-to-speech (NAR-TTS), containing official implementations of PortaSpeech (NeurIPS 2021) and … | 23 | 1004 | maintenance |
| LeeSureman/Flat-Lattice-Transformer Reference implementation of the ACL 2020 paper FLAT: Chinese NER Using Flat-Lattice Transformer, built on PyTorch and FastNLP. It trains fl… | 32 | 1002 | maintenance |
| google-research/magvit Official JAX implementation of MAGVIT, a masked generative video transformer from a CVPR 2023 paper by Google Research and CMU. It provides… | 10 | 1002 | maintenance |
| omertov/encoder4editing Official PyTorch implementation of the SIGGRAPH 2021 paper 'Designing an Encoder for StyleGAN Image Manipulation' (e4e). It encodes real im… | 32 | 1000 | maintenance |
| openai/grok Research code accompanying the paper 'Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets' from OpenAI. It provides t… | 10 | 4259 | experimental |
| 0hq/WebGPT WebGPT is a vanilla JavaScript and HTML implementation of GPT transformer inference running in the browser via WebGPU, in under ~1500 lines… | 30 | 3787 | experimental |
| MoonshotAI/Attention-Residuals Official implementation of Attention Residuals (AttnRes), a drop-in replacement for standard residual connections in Transformers that lets… | 44 | 3516 | experimental |
| lucidrains/musiclm-pytorch A PyTorch library implementing MusicLM, Google's text-to-music generation model, by combining text-conditioned AudioLM with MuLan, a text-a… | 22 | 3292 | experimental |
| enhuiz/vall-e An unofficial PyTorch implementation of the VALL-E text-to-speech audio language model, built on the EnCodec tokenizer. It provides trainin… | 32 | 2967 | experimental |
| guoqincode/Open-AnimateAnyone An unofficial PyTorch implementation of Animate Anyone, a diffusion-based method that animates a static character image using pose sequence… | 27 | 2917 | experimental |
| Everlyn-Labs/Everlyn-1 Everlyn-1 is an open autoregressive foundational video AI model from Everlyn Labs, accompanied by research on video compression/tokenizatio… | 67 | 2893 | experimental |
| test-time-training/ttt-video-dit Official PyTorch implementation of 'One-Minute Video Generation with Test-Time Training', which finetunes the CogVideoX 5B diffusion transf… | 48 | 2451 | experimental |
| dvmazur/mixtral-offloading A Python library enabling efficient inference of Mixtral-8x7B mixture-of-experts language models on limited hardware like Google Colab or c… | 27 | 2334 | experimental |
| google/tangent Tangent is a Python library for automatic differentiation that works by source-to-source transformation: it reads your Python function's so… | 10 | 2326 | experimental |
| etched-ai/open-oasis Inference code and model weights for Oasis 500M, an interactive world model from Decart and Etched that generates gameplay video autoregres… | 22 | 2145 | experimental |
| JiauZhang/DragGAN A Python implementation of DragGAN, a research method for interactively manipulating generated images by dragging points on the generative … | 30 | 2123 | experimental |
| lucidrains/make-a-video-pytorch A PyTorch library implementing Make-A-Video, Meta AI's text-to-video generation approach, built around pseudo-3d (axial) convolutions and s… | 23 | 1989 | experimental |
| lucidrains/gigagan-pytorch A PyTorch implementation of GigaGAN, Adobe's state-of-the-art generative adversarial network for text-to-image and unconditional image synt… | 22 | 1945 | experimental |