Open-source machine-learning software
5374 open-source libraries, tools and applications tagged machine-learning, ranked by how closely machine-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.
5374 products, primary matches first, then adoption-weighted; health v2 shown.
| Product | Health v2 | Stars | Maturity |
|---|---|---|---|
| nihui/waifu2x-ncnn-vulkan A portable command-line tool implementing the waifu2x anime-style image upscaler and denoiser using the ncnn inference framework with the V… | 57 | 3474 | active |
| TextAttack TextAttack is a Python framework for generating adversarial examples against NLP models, as well as for data augmentation and model trainin… | 80 | 3471 | active |
| deepseek-ai/DeepSeek-OCR-2 DeepSeek-OCR 2 is an open-source vision-language model and inference toolkit implementing 'Visual Causal Flow' for optical character recogn… | 42 | 3471 | active |
| MiniMax AI Platform Official repository for MiniMax-Text-01 and MiniMax-VL-01, open-weight large language and vision-language models built on a linear attentio… | 31 | 3469 | active |
| lakehq/sail Sail is an open-source, Rust-native multimodal compute engine that serves as a drop-in replacement for Apache Spark, unifying batch process… | 94 | 3465 | active |
| shenweichen/DeepCTR-Torch DeepCTR-Torch is a PyTorch library providing easy-to-use, modular, and extendable implementations of deep-learning-based CTR (click-through… | 71 | 3460 | active |
| aiming-lab/MetaClaw MetaClaw is a continual meta-learning framework that lets an LLM agent evolve from real conversations, combining skill synthesis from failu… | 63 | 3460 | active |
| PaddlePaddle/PARL PARL is a flexible, high-performance reinforcement learning framework built on PaddlePaddle, providing Model/Algorithm/Agent abstractions a… | 38 | 3457 | active |
| MTEB MTEB (Massive Text Embedding Benchmark) is a Python library and CLI for benchmarking embedding models across 1000+ tasks, languages, and mo… | 95 | 3455 | active |
| modelscope/evalscope EvalScope is a Python framework from the ModelScope community for evaluating large language models, vision-language models, embedding model… | 91 | 3454 | active |
| XinJingHao/DRL-Pytorch A unified PyTorch implementation collection of popular deep reinforcement learning algorithms including DQN variants, PPO, DDPG, TD3, SAC, … | 40 | 3447 | active |
| mani-skill/ManiSkill ManiSkill is an open-source GPU-parallelized robotics simulation framework and benchmark built on SAPIEN, focused on manipulation skills. I… | 85 | 3442 | active |
| aqlaboratory/openfold OpenFold is a faithful, trainable PyTorch reproduction of DeepMind's AlphaFold 2 for protein structure prediction. It is memory-efficient a… | 45 | 3435 | active |
| luigifreda/pyslam pySLAM is a hybrid Python/C++ Visual SLAM pipeline supporting monocular, stereo, and RGB-D cameras with a wide range of local and global fe… | 76 | 3430 | active |
| NVlabs/stylegan The official TensorFlow implementation of StyleGAN, NVIDIA's style-based generator architecture for generative adversarial networks from th… | 32 | 14414 | maintenance |
| AnimateDiff for Stable Diffusion WebUI An AUTOMATIC1111 Stable Diffusion WebUI extension that integrates AnimateDiff motion modules to generate animated GIFs and videos from diff… | 29 | 3425 | active |
| RKNN-Toolkit2 RKNN-Toolkit2 is Rockchip's SDK for converting trained neural network models into RKNN format and deploying them on Rockchip NPU chips like… | 33 | 3419 | active |
| microsoft/nni NNI (Neural Network Intelligence) is an open-source AutoML toolkit from Microsoft that automates hyperparameter tuning, neural architecture… | 10 | 14367 | maintenance |
| davidsandberg/facenet A TensorFlow implementation of the FaceNet face recognizer that generates 128-dimensional face embeddings, including face detection via MTC… | 32 | 14357 | maintenance |
| jsvine/markovify Markovify is a simple, extensible Python library for building Markov chain models from text corpora and generating random semi-plausible se… | 32 | 3409 | stable |
| IQA-PyTorch A pure Python/PyTorch toolbox for image quality assessment (IQA) providing GPU-accelerated reimplementations of many full-reference and no-… | 82 | 3408 | active |
| HKUDS/VideoRAG VideoRAG is a retrieval-augmented generation framework for chatting with and understanding extremely long-context videos, using a dual-chan… | 50 | 3401 | active |
| NovaSky-AI/SkyThought SkyThought is the open-source repository behind Sky-T1, a family of reasoning language models trained for under $450, including training sc… | 22 | 3401 | active |
| opengeos/geoai GeoAI is a Python package that integrates artificial intelligence with geospatial data analysis, built on PyTorch, Transformers, and segmen… | 89 | 3390 | active |
| magenta/ddsp DDSP is a Python library of differentiable digital signal processing components (synthesizers, filters, waveshapers) that can be embedded i… | 68 | 3380 | active |
| CompVis/latent-diffusion The official research code and pretrained model zoo for Latent Diffusion Models (LDM), the paper behind Stable Diffusion, enabling high-res… | 32 | 14168 | maintenance |
| WongKinYiu/yolov7 Official PyTorch implementation of the YOLOv7 paper, a state-of-the-art real-time object detector with trainable bag-of-freebies techniques… | 23 | 14153 | maintenance |
| PhysicsNeMo NVIDIA PhysicsNeMo is an open-source Python deep-learning framework for building, training, fine-tuning, and inferring physics AI models us… | 93 | 3373 | active |
| libAudioFlux/audioFlux audioFlux is a C-based library with Python bindings for audio and music analysis and feature extraction. It supports dozens of time-frequen… | 50 | 3372 | active |
| VainF/Torch-Pruning Torch-Pruning is a PyTorch framework for structural neural network pruning based on the DepGraph algorithm from CVPR 2023. It automatically… | 44 | 3359 | active |
| shankarpandala/lazypredict Lazy Predict is a Python library that trains dozens of machine learning models with minimal code to quickly identify which algorithms perfo… | 86 | 3352 | active |
| Lagrange-Labs/deep-prove DeepProve is a Rust framework for generating zero-knowledge proofs of neural network inference, including end-to-end proving of full LLM fo… | 57 | 3352 | active |
| ltdrdata/ComfyUI-Impact-Pack A custom node pack for ComfyUI that enhances Stable Diffusion image generation through detectors, detailers, upscalers, and pipeline utilit… | 62 | 3346 | active |
| xenova/whisper-web A browser-based speech recognition app that runs OpenAI's Whisper models entirely client-side using Transformers.js. It transcribes audio w… | 31 | 3345 | active |
| pgmpy pgmpy is a Python library for causal and probabilistic reasoning with graphical models such as Bayesian Networks, Dynamic Bayesian Networks… | 78 | 3338 | stable |
| ResearAI/DeepScientist DeepScientist is a local-first autonomous research studio that runs the full scientific research loop on your machine, from baselines and e… | 68 | 3335 | active |
| Peterande/D-FINE D-FINE is the official PyTorch implementation of an ICLR 2025 Spotlight paper that redefines the regression task in DETR-style detectors as… | 64 | 3334 | active |
| Opus libopus is the reference implementation of the Opus audio codec, an IETF-standardized (RFC 6716) royalty-free codec for interactive speech … | 65 | 3330 | stable |
| Bytez-com/docs Bytez is a serverless model inference platform offering a unified API with a single key to run 175k+ open and closed-source AI models (LLMs… | 53 | 3329 | active |
| Farama-Foundation/HighwayEnv HighwayEnv is a collection of Gymnasium environments for autonomous driving and tactical decision-making tasks, covering scenarios like hig… | 92 | 3325 | active |
| vladmandic/human Human is a JavaScript/TypeScript library built on TensorFlow.js that combines multiple ML models for 3D face detection and recognition, bod… | 44 | 3324 | active |
| LLM-as-a-Verifier A Python framework that uses LLM token logprobs to produce fine-grained verification scores for agent trajectories without additional train… | 59 | 3322 | active |
| microsoft/LoRA loralib is the official PyTorch implementation of LoRA (Low-Rank Adaptation), which fine-tunes large language models by injecting trainable… | 23 | 13868 | maintenance |
| 2D Gaussian Splatting Official implementation of 2D Gaussian Splatting (2DGS), a SIGGRAPH 2024 method that represents scenes as 2D oriented Gaussian disks for ge… | 66 | 3312 | stable |
| mljar/mljar-supervised MLJAR AutoML is a Python package for automated machine learning on tabular data, providing feature engineering, hyperparameter tuning, mode… | 85 | 3294 | active |
| facebookresearch/tribev2 TRIBE v2 is a multimodal deep learning model from Meta AI that predicts fMRI brain responses to naturalistic video, audio, and text stimuli… | 51 | 3293 | active |
| timerring/bilive BILIVE is a Python application that records Bilibili live streams and danmaku 24/7, then automatically renders danmaku and AI-generated sub… | 57 | 3291 | active |
| thomasahle/sunfish Sunfish is a simple but strong chess engine written in Python in roughly 111 lines of code, using MTD-bi search with piece-square table eva… | 93 | 3289 | active |
| Tencent-Hunyuan/HunyuanImage-3.0 HunyuanImage-3.0 is Tencent's open-source native multimodal model for text-to-image and image-to-image generation, with inference code and … | 54 | 3279 | active |
| jixiaozhong/Sonic Sonic is the official PyTorch implementation of the CVPR 2025 paper 'Sonic: Shifting Focus to Global Audio Perception in Portrait Animation… | 46 | 3273 | active |
| MAIF/shapash Shapash is a Python library that makes machine learning models interpretable and understandable through clear visualizations, a webapp, and… | 95 | 3267 | active |
| deepdoctection deepdoctection is a Python library for Document AI that orchestrates document layout analysis, table recognition, OCR, and document/token c… | 96 | 3261 | active |
| cocktailpeanut/fluxgym FluxGym is a simple web UI for training FLUX LoRA models with low VRAM support (12GB/16GB/20GB). It combines the AI-Toolkit Gradio frontend… | 62 | 3257 | active |
| LeelaChessZero/lc0 Lc0 is an open-source, UCI-compliant chess engine that plays chess using neural networks trained via AlphaZero-style self-play reinforcemen… | 74 | 3248 | active |
| Beckschen/TransUNet Official PyTorch implementation of TransUNet, a U-Net-style architecture that uses a Vision Transformer encoder for medical image segmentat… | 60 | 3245 | stable |
| determined-ai/determined Determined is an open-source deep learning platform that combines distributed training, hyperparameter tuning, experiment tracking, and GPU… | 36 | 3245 | active |
| Nerogar/OneTrainer OneTrainer is a GUI and CLI application for fine-tuning diffusion image models, supporting full fine-tuning, LoRA, and embeddings across ma… | 75 | 3243 | active |
| MisoLabsAI/MisoTTS Miso TTS 8B is an open-source text-to-speech model based on an RVQ Transformer architecture with a Llama 3.2-style 8B backbone, designed fo… | 50 | 3240 | active |
| onnx/onnx-tensorrt A C++ parser library and backend that converts ONNX models into TensorRT engines for high-performance GPU inference. It is maintained by NV… | 91 | 3238 | active |
| Jittor/jittor Jittor is a high-performance deep learning framework from Tsinghua University based on just-in-time (JIT) compilation and meta-operators, w… | 68 | 3237 | active |
| Pointcept Pointcept is a PyTorch-based research codebase for point cloud perception, providing implementations of state-of-the-art 3D scene understan… | 74 | 3233 | active |
| prs-eth/Marigold Marigold is a family of diffusion-based models and a fine-tuning protocol that adapts pretrained latent diffusion models like Stable Diffus… | 69 | 3233 | active |
| mit-han-lab/bevfusion BEVFusion is a PyTorch-based multi-task multi-sensor fusion framework that unifies camera and LiDAR features in a shared bird's-eye view re… | 10 | 3231 | stable |
| VMamba VMamba is a PyTorch implementation of a visual state space model (SSM) vision backbone based on Mamba, featuring 2D Selective Scan (SS2D) f… | 18 | 3230 | active |
| Mayandev/notion-avatar Notion Avatar Maker is an AI-powered web application for creating Notion-style illustrated avatars from photos or text descriptions, with a… | 67 | 3227 | active |
| facebookresearch/dinov2 PyTorch implementation and pretrained models for DINOv2, a self-supervised vision transformer method from Meta AI that learns robust visual… | 75 | 13422 | maintenance |
| xianfei/SysMocap SysMocap is a cross-platform, video-driven real-time motion capture system that animates 3D virtual characters from webcam footage. It rend… | 72 | 3223 | active |
| echohive42/AI-reads-books-page-by-page A Python script that reads PDF books page by page, using AI to extract knowledge points and generate progressive summaries at configurable … | 58 | 3217 | active |
| qdrant/fastembed FastEmbed is a lightweight Python library for generating text embeddings using ONNX Runtime instead of PyTorch, requiring no GPU and minima… | 90 | 3216 | active |
| Captcha Solution A deep learning training tool for image CAPTCHA recognition built on TensorFlow, using CNN/ResNet/DenseNet backbones with GRU/LSTM recurren… | 52 | 3215 | active |
| jy0205/Pyramid-Flow Pyramid Flow is the official PyTorch implementation of a training-efficient autoregressive video generation model based on pyramidal flow m… | 23 | 3213 | active |
| Djdefrag/QualityScaler QualityScaler is a Windows GUI application that uses AI deep-learning models to upscale, enhance, and de-noise images and videos. It is wri… | 89 | 3207 | active |
| megvii-research/NAFNet NAFNet is the official PyTorch implementation of a state-of-the-art image restoration network that removes nonlinear activation functions. … | 32 | 3205 | stable |
| Arm NN Arm's Compute Library is a C++ collection of over 100 low-level machine learning and computer vision functions optimized for Arm Cortex-A/N… | 98 | 3197 | active |
| rmurai0610/MASt3R-SLAM MASt3R-SLAM is a real-time monocular dense SLAM system built on the MASt3R two-view 3D reconstruction prior, producing globally consistent … | 40 | 3191 | active |
| imbue-bit/AlphaGPT AlphaGPT is an open-source automated factor factory based on deep reinforcement learning for quantitative finance. It mines and generates a… | 52 | 3187 | active |
| Miles Miles is an open-source, enterprise-grade reinforcement learning framework for large-scale LLM and VLM post-training, forked from and co-ev… | 84 | 3185 | active |
| stepfun-ai/Step-Video-T2V Step-Video-T2V is an open-source text-to-video generation model from StepFun, released with inference code and pretrained weights (includin… | 22 | 3185 | active |
| Rudrabha/Wav2Lip Wav2Lip is the official research code for the ACM Multimedia 2020 paper 'A Lip Sync Expert Is All You Need for Speech to Lip Generation In … | 41 | 13221 | maintenance |
| MiniMax AI Platform MiniMax-M1 is an open-weight, large-scale hybrid-attention reasoning language model released by MiniMax under Apache-2.0. The repository pr… | 29 | 3184 | active |
| tslearn-team/tslearn tslearn is a Python machine learning library dedicated to time series analysis, built on numpy and compatible with scikit-learn. It provide… | 84 | 3183 | stable |
| tekaratzas/RustGPT A transformer-based large language model implemented entirely in pure Rust with no external ML frameworks, using only ndarray for matrix op… | 35 | 3175 | active |
| stemrollerapp/stemroller StemRoller is a free open-source desktop app that separates vocals, drums, bass, and other stems from any song with a single click, using F… | 77 | 3173 | active |
| mjlab mjlab is a Python framework that combines Isaac Lab's manager-based API with MuJoCo Warp, a GPU-accelerated version of MuJoCo, for reinforc… | 82 | 3163 | active |
| nesaorg/nesa Nesa is a privacy-preserving AI inference platform that runs models like Llama, Mistral, and Stable Diffusion with end-to-end encryption us… | 22 | 3163 | active |
| SciML/DifferentialEquations.jl A Julia suite of high-performance numerical solvers for differential equations, covering ODEs, SDEs, DDEs, DAEs, RODEs, jumps, and (S)PDEs,… | 90 | 3162 | active |
| parrt/dtreeviz dtreeviz is a Python library for visualizing decision trees and interpreting tree-based machine learning models. It supports scikit-learn, … | 63 | 3157 | active |
| scikit-learn-contrib/hdbscan A high-performance Python implementation of the HDBSCAN hierarchical density-based clustering algorithm, part of the scikit-learn-contrib e… | 88 | 3154 | stable |
| ali-vilab/VGen VGen is the official repository for a holistic video generation ecosystem built on diffusion models, including the I2VGen-XL cascaded image… | 28 | 3154 | active |
| cleardusk/3DDFA_V2 3DDFA_V2 is the official PyTorch implementation of the ECCV 2020 paper 'Towards Fast, Accurate and Stable 3D Dense Face Alignment'. It regr… | 23 | 3150 | stable |
| modelscope/3D-Speaker 3D-Speaker is an open-source Python toolkit for single- and multi-modal speaker verification, speaker recognition, and speaker diarization,… | 53 | 3148 | active |
| dexmal/dexbotic Dexbotic is an open-source PyTorch-based toolbox for developing Vision-Language-Action (VLA) models for embodied intelligence. It unifies p… | 70 | 3144 | active |
| thuml/Time-Series-Library TSLib is an open-source Python library providing a unified codebase of advanced deep learning models for general time series analysis. It s… | 62 | 12923 | maintenance |
| naver/mast3r MASt3R is the official PyTorch implementation of 'Grounding Image Matching in 3D with MASt3R' (ECCV 2024), a model that performs dense 3D r… | 34 | 3114 | active |
| junyanz/CycleGAN A Torch (Lua) implementation of CycleGAN and pix2pix for unpaired image-to-image translation using cycle-consistent adversarial networks. I… | 32 | 12874 | maintenance |
| Top2Vec Top2Vec is a Python library that learns jointly embedded topic, document, and word vectors for topic modeling and semantic search. It suppo… | 23 | 3111 | stable |
| AutoArk/GPA GPA (General Purpose Audio) is a unified autoregressive audio-language model that performs text-to-speech, automatic speech recognition, an… | 51 | 3109 | active |
| jina-ai/clip-as-service CLIP-as-service is a low-latency, high-scalability server for embedding images and text into fixed-length vectors using OpenAI's CLIP model… | 23 | 12831 | maintenance |
| TensorFlow Lite for Microcontrollers TensorFlow Lite for Microcontrollers (TFLM) is a C++ port of TensorFlow Lite for running ML models on microcontrollers, DSPs, and other mem… | 76 | 3098 | active |
| ridgerchu/matmulfreellm A Python implementation of MatMul-Free LM, a language model architecture that eliminates matrix multiplication operations using ternary wei… | 76 | 3093 | active |