Open-source machine-learning software
4007 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.
4007 products, primary matches first, then adoption-weighted; health v2 shown.
| Product | Health v2 | Stars | Maturity |
|---|---|---|---|
| rzru/nightingale Nightingale is an open-source karaoke application that turns any song in your music library into a karaoke track using neural networks. It … | 79 | 1475 | active |
| Lightning-AI/lightning-thunder Lightning Thunder is a source-to-source deep learning compiler for PyTorch that optimizes models for training and inference. It provides a … | 70 | 1475 | active |
| Om-Alve/smolGPT A minimal pure-PyTorch implementation for training small GPT-style LLMs from scratch, featuring flash attention, RMSNorm, SwiGLU, RoPE, and… | 21 | 1474 | active |
| hojonathanho/diffusion The official reference implementation of Denoising Diffusion Probabilistic Models (DDPM) from the 2020 paper by Jonathan Ho et al., written… | 32 | 5341 | maintenance |
| refinery An open-source tool for scaling, assessing, and maintaining natural language training data, treating datasets like software artifacts. It p… | 23 | 1473 | active |
| postgresml/korvus Korvus is an open-source search SDK that unifies the entire RAG pipeline—embedding generation, vector search, reranking, summarization, and… | 24 | 1470 | active |
| LocoMuJoCo LocoMuJoCo is an imitation learning benchmark for whole-body locomotion control built on MuJoCo, featuring humanoid, quadruped, and musculo… | 70 | 1468 | active |
| jax-md/jax-md JAX MD is a Python library for molecular dynamics simulations built on JAX, making them hardware accelerated on CPU, GPU, and TPU and end-t… | 80 | 1465 | active |
| facebookresearch/MobileLLM Meta's training code for MobileLLM, a family of sub-billion parameter language models optimized for on-device use, published at ICML 2024. … | 56 | 1465 | active |
| FeiYull/TensorRT-Alpha A C++/CUDA library providing TensorRT-accelerated deployment for 30+ popular computer vision models including YOLOv3-v8, YOLOv8-Pose/Seg/Cl… | 100 | 1464 | active |
| tianweiy/DMD2 DMD2 is the official PyTorch implementation of Improved Distribution Matching Distillation, a NeurIPS 2024 method that distills diffusion m… | 25 | 1464 | active |
| yunjey/stargan Official PyTorch implementation of StarGAN, a unified generative adversarial network for multi-domain image-to-image translation (CVPR 2018… | 32 | 5295 | maintenance |
| microsoft/NeuralSpeech NeuralSpeech is a Microsoft Research Asia research repository containing implementations of neural speech processing models across ASR erro… | 10 | 1461 | active |
| Turing-Project/WriteGPT WriteGPT is a generative text-creation AI framework built on GPT-2 and other models (EAST, CRNN, BERT), fine-tuned to generate Chinese exam… | 23 | 5282 | maintenance |
| pyradiomics PyRadiomics is an open-source Python package for extracting radiomics features from 2D and 3D medical images and binary masks. It provides … | 41 | 1460 | active |
| tensorflow/tpu A collection of reference models and tools for training machine learning models on Google Cloud TPUs, maintained as a public mirror by the … | 68 | 5278 | maintenance |
| ByteDance-Seed/m3-agent M3-Agent is a multimodal agent framework from ByteDance Seed that processes real-time visual and auditory inputs to build entity-centric lo… | 45 | 1459 | active |
| aeon-toolkit/aeon aeon is a scikit-learn compatible Python toolkit for machine learning on time series, covering classification, regression, clustering, fore… | 93 | 1457 | active |
| sdcb/PaddleSharp A .NET/C# wrapper around Baidu's PaddleInference C API, providing PaddleOCR, PaddleDetection, rotation detection, Chinese segmentation, and… | 75 | 1453 | active |
| zama-ai/concrete-ml Concrete ML is a privacy-preserving machine learning library built on top of Zama's Concrete FHE compiler. It lets data scientists convert … | 65 | 1453 | active |
| YanjieZe/3D-Diffusion-Policy 3D Diffusion Policy (DP3) is a visual imitation learning algorithm that combines compact 3D point cloud representations with diffusion poli… | 43 | 1453 | active |
| dbolya/yolact YOLACT is a PyTorch implementation of a fully convolutional model for real-time instance segmentation, accompanying the YOLACT and YOLACT++… | 47 | 5242 | maintenance |
| dleemiller/WordLlama WordLlama is a fast, lightweight Python NLP toolkit built on LLM token embeddings for tasks like similarity computation, ranking, fuzzy ded… | 42 | 1450 | active |
| google-deepmind/rlax RLax is a JAX-based library of building blocks for implementing reinforcement learning agents, providing mathematical operations like value… | 79 | 1449 | active |
| microsoft/KBLaM Official implementation of KBLaM, a method for augmenting pre-trained LLMs with external knowledge by encoding a knowledge base into contin… | 68 | 1449 | active |
| zylo117/Yet-Another-EfficientDet-Pytorch A PyTorch re-implementation of Google's EfficientDet object detection model with pretrained weights matching near-SOTA accuracy at real-tim… | 23 | 5233 | maintenance |
| ThoughtfulDev/EagleEye EagleEye is a Python-based OSINT tool that identifies social media profiles (Instagram, Facebook, Twitter, YouTube) of a person using face … | 32 | 5231 | maintenance |
| amdegroot/ssd.pytorch A PyTorch implementation of the Single Shot MultiBox Detector (SSD) object detection model from the 2016 paper by Wei Liu et al. It include… | 32 | 5222 | maintenance |
| SociallyIneptWeeb/AICoverGen AICoverGen is a Python WebUI (with CLI support) that generates AI song covers by converting vocals in a song to any RVC v2 trained voice, s… | 29 | 1445 | active |
| deepseek-ai/EPLB EPLB is DeepSeek's open-source Expert Parallelism Load Balancer for Mixture-of-Experts models. It computes balanced expert replication and … | 23 | 1443 | active |
| CliMA Earth System Model Oceananigans.jl is a Julia package for fast, flexible finite-volume simulations of incompressible fluid dynamics, solving nonhydrostatic an… | 96 | 1442 | active |
| SakanaAI/evolutionary-model-merge Official repository for SakanaAI's Evolutionary Model Merge research, providing code and resources to reproduce paper evaluations of models… | 16 | 1442 | active |
| Agent-RL/ReCall ReCall is a framework that trains LLMs to reason with arbitrary tool calls via reinforcement learning, without supervised tool-use trajecto… | 26 | 1441 | active |
| tensorspace-team/tensorspace TensorSpace is a neural network 3D visualization framework built on TensorFlow.js, Three.js, and Tween.js. It provides Keras-like APIs to b… | 23 | 5198 | maintenance |
| nilearn/nilearn Nilearn is a Python library providing statistical and machine-learning tools for analyzing brain imaging data such as fMRI and MRI volumes … | 91 | 1439 | active |
| sql-machine-learning/sqlflow SQLFlow is a compiler that extends SQL with AI-oriented syntax (training, prediction, evaluation, explanation, and mathematical programming… | 23 | 5190 | maintenance |
| voice-cloning-app/Voice-Cloning-App A Python/PyTorch desktop application for cloning and synthesizing human voices from audio datasets. It handles the full pipeline from autom… | 23 | 1438 | active |
| edwko/OuteTTS OuteTTS is a Python interface for running OuteAI's text-to-speech models, supporting multiple inference backends including llama.cpp, Huggi… | 49 | 1437 | active |
| zuruoke/watermark-removal A machine learning tool that removes watermarks from images using deep learning image inpainting, based on Contextual Attention and Gated C… | 77 | 5181 | maintenance |
| JuliaGPU/CUDA.jl CUDA.jl is the main Julia package for programming NVIDIA CUDA GPUs, offering a high-level CuArray abstraction, a compiler for writing CUDA … | 99 | 1433 | stable |
| ymcui/Chinese-ELECTRA Chinese-ELECTRA is a collection of pre-trained Chinese ELECTRA language models released by the HIT-iFLYTEK joint lab, built on the official… | 63 | 1433 | stable |
| kitops-ml/kitops KitOps is a CNCF open-source DevOps tool (Kit CLI, written in Go) that packages AI/ML models, datasets, code, prompts, agent skills, and MC… | 92 | 1430 | active |
| zsyOAOA/ResShift ResShift is an efficient diffusion model for image super-resolution that transfers between low- and high-resolution images by shifting resi… | 59 | 1430 | active |
| facebookresearch/vggsfm VGGSfM is a deep learning-based Structure from Motion pipeline from Meta AI and Oxford VGG that recovers camera poses and 3D point clouds f… | 26 | 1430 | active |
| jasonmayes/Real-Time-Person-Removal A browser-based demo that removes people from complex video backgrounds in real time using TensorFlow.js. It learns the static background o… | 23 | 5151 | maintenance |
| winedarksea/AutoTS AutoTS is a Python library for automated time series forecasting, offering dozens of sklearn-style models (statistical, ML, deep learning) … | 88 | 1429 | active |
| HongyuanLuke/frequencylaw Official code repository for the paper 'Textual Frequency Law on Large Language Models', implementing TFL, TFD, and CTFT methods for studyi… | 46 | 1428 | active |
| kootenpv/whereami A Python CLI tool that uses WiFi access point signal strengths and a scikit-learn RandomForest model to predict your indoor location. It le… | 32 | 5142 | maintenance |
| wenet-e2e/wespeaker WeSpeaker is a research and production-oriented toolkit for speaker embedding learning, supporting speaker verification, recognition, and d… | 66 | 1427 | active |
| argilla-io/argilla Argilla is an open-source collaboration tool for AI engineers and domain experts to build, annotate, and curate high-quality datasets for N… | 78 | 5140 | maintenance |
| roshan-research/hazm Hazm is a Python library for natural language processing on Persian (Farsi) text, offering normalization, tokenization, stemming, lemmatiza… | 70 | 1426 | active |
| apple/ml-aim Apple's official repository for AIM (Autoregressive Image Models), providing code and pretrained checkpoints for AIMv1 and AIMv2 large visi… | 69 | 1426 | active |
| mdbloice/Augmentor Augmentor is a standalone Python library for image augmentation in machine learning, providing a pipeline of stochastic operations like rot… | 32 | 5131 | maintenance |
| ZheC/Realtime_Multi-Person_Pose_Estimation Reference implementation of the CVPR'17 paper 'Realtime Multi-Person Pose Estimation', a bottom-up approach that detects keypoints for mult… | 32 | 5126 | maintenance |
| SagiPolaczek/NeuralSVG Official PyTorch implementation of NeuralSVG, an ICCV 2025 paper that generates layered, editable SVG vector graphics from text prompts. It… | 44 | 1421 | active |
| lyst/lightfm LightFM is a Python library implementing hybrid recommendation algorithms that combine collaborative filtering with user and item metadata … | 23 | 5116 | maintenance |
| MLOps Python Package A Python package template that provides a production-grade code base with MLOps best practices for building and deploying machine learning … | 92 | 1419 | active |
| HugoTini/DeepBump DeepBump is a machine-learning tool that generates normal, height, and curvature maps from single pictures, using a U-Net (MobileNetV2) mod… | 24 | 1419 | active |
| oracle/tribuo Tribuo is a Java machine learning library from Oracle Labs providing classification, regression, clustering, anomaly detection, and multi-l… | 55 | 1418 | active |
| NVlabs/BundleSDF BundleSDF is a CVPR 2023 research implementation from NVIDIA for near real-time 6-DoF pose tracking of unknown rigid objects from monocular… | 62 | 1417 | stable |
| ARahim3/mlx-tune A Python library for fine-tuning LLMs, vision-language, audio (TTS/STT), embedding, OCR, and JEPA models natively on Apple Silicon Macs usi… | 69 | 1416 | active |
| jrzaurin/pytorch-widedeep A PyTorch library for multimodal deep learning that combines tabular data with text and images using Wide and Deep model architectures. It … | 55 | 1416 | active |
| mratsim/Arraymancer Arraymancer is a fast, ergonomic N-dimensional tensor (ndarray) library written in Nim, inspired by NumPy and PyTorch. It provides CPU, CUD… | 67 | 1415 | active |
| Terrain Diffusion Terrain Diffusion is a Python framework that uses diffusion models as a learned, deterministic replacement for Perlin noise, generating inf… | 58 | 1414 | active |
| Lyken17/pytorch-OpCounter THOP (PyTorch-OpCounter) is a Python library that counts the MACs/FLOPs and parameters of PyTorch models. It profiles arbitrary nn.Modules … | 32 | 5083 | maintenance |
| explosion/spacy-transformers A spaCy v3 extension package that provides pipeline components for using pretrained transformer models like BERT, RoBERTa, XLNet, and GPT-2… | 65 | 1413 | stable |
| CSAILVision/semantic-segmentation-pytorch A PyTorch implementation of semantic segmentation (scene parsing) models for the MIT ADE20K dataset, including pretrained model zoo and tra… | 32 | 5081 | maintenance |
| affinelayer/pix2pix-tensorflow A TensorFlow implementation of pix2pix, a conditional GAN that learns a mapping from input images to output images. It is a faithful port o… | 32 | 5078 | maintenance |
| koaning/scikit-lego scikit-lego is a Python library of extra building blocks—custom transformers, models, metrics, and datasets—that are fully compatible with … | 88 | 1412 | active |
| lucidrains/self-rewarding-lm-pytorch A PyTorch library implementing the Self-Rewarding Language Model training framework from MetaAI, along with the SPIN training method. It pr… | 17 | 1412 | active |
| nyrahealth/CrisperWhisper CrisperWhisper 2.0 is a controllable speech recognition model and Python library that transcribes audio either verbatim (including fillers,… | 88 | 1411 | active |
| scraed/LanPaint LanPaint is a training-free inpainting sampler for stable diffusion models, implemented as a ComfyUI custom node. It uses iterative 'think … | 83 | 1409 | active |
| PrimiHub PrimiHub is an open-source privacy-preserving computing platform built by a team of cryptography experts, supporting secure multi-party com… | 40 | 1409 | active |
| lucidrains/transfusion-pytorch A PyTorch implementation of Transfusion, MetaAI's approach to predicting the next token and diffusing images with a single multi-modal mode… | 69 | 1408 | active |
| om-ai-lab/OmDet OmDet-Turbo is a PyTorch implementation of a transformer-based open-vocabulary object detection model that detects arbitrary user-defined o… | 54 | 1406 | active |
| open-gigaai/giga-world-policy GigaWorld-Policy is a World Action Model (WAM) for robot policy learning that jointly models actions and future visual observations during … | 53 | 1406 | active |
| XiaoMi/mace MACE (Mobile AI Compute Engine) is a deep learning inference framework optimized for mobile heterogeneous computing on Android, iOS, Linux … | 23 | 5052 | maintenance |
| Irodori-TTS Irodori-TTS is a Flow Matching-based text-to-speech model with training and inference code, built on a Rectified Flow Diffusion Transformer… | 58 | 1401 | active |
| RahulSChand/gpu_poor A web-based calculator that estimates GPU memory requirements and inference/finetuning throughput (token/s) for any LLM. It supports quanti… | 19 | 1401 | active |
| McGill-NLP/webllama WebLlama is a framework for building Llama-3-powered agents that browse the web by following natural language instructions and dialogue. It… | 56 | 1399 | active |
| Denys88/rl_games RL Games is a high-performance reinforcement learning library built on PyTorch, focused on training agents in massively parallel GPU-based … | 74 | 1397 | active |
| explosion/spacy-llm spacy-llm is a Python library that integrates Large Language Models into spaCy NLP pipelines via a serializable llm component. It provides … | 75 | 1395 | active |
| xl0/lovely-tensors A Python library that provides human-friendly summaries of PyTorch tensors, showing shape, statistics, histograms, and NaN/Inf detection in… | 63 | 1395 | active |
| TheRock TheRock is AMD's open-source build and release system for the ROCm software stack, replacing the legacy monolithic ROCm release process wit… | 89 | 1394 | active |
| galilai-group/lejepa LeJEPA is a Python framework for scalable, theoretically grounded self-supervised representation learning based on Joint-Embedding Predicti… | 42 | 1393 | active |
| Junyi42/monst3r MonST3R is the official PyTorch implementation of an ICLR 2025 paper that estimates per-timestep geometry (pointmaps) from dynamic videos i… | 22 | 1390 | active |
| PyO3/rust-numpy Rust bindings for the NumPy C-API built on PyO3, enabling zero-copy conversion between Rust ndarray types and NumPy arrays. It supports wri… | 85 | 1388 | active |
| QwenLM/Qwen3-VL-Embedding Qwen3-VL-Embedding and Qwen3-VL-Reranker are state-of-the-art multimodal embedding and reranking models built on the Qwen3-VL foundation mo… | 52 | 1387 | active |
| haoheliu/voicefixer VoiceFixer is a Python library and CLI tool for general speech restoration, using a pretrained neural vocoder to restore degraded human spe… | 23 | 1387 | stable |
| tiny-tpu-v2/tiny-tpu A minimal tensor processing unit (TPU) written in Verilog, reinvented from Google's TPU V1/V2 architectures. It includes a systolic array o… | 50 | 1386 | active |
| zjunlp/KnowLM KnowLM is an open-source framework for building knowledgeable large language models, covering data processing, pre-training, fine-tuning, k… | 31 | 1386 | active |
| meridianlabs-ai/inspect_petri Inspect Petri is an alignment auditing agent that automatically probes language models for concerning behaviors like sycophancy, reward hac… | 81 | 1385 | active |
| lorey/mlscraper mlscraper is a Python library that automatically extracts structured data from HTML pages using machine learning. Instead of writing CSS se… | 23 | 1385 | active |
| soda-inria/tabicl TabICLv2 is an open-source tabular foundation model that performs classification and regression via a single forward pass through a pre-tra… | 83 | 1384 | active |
| Zefan-Cai/KVCache-Factory KVCache-Factory is a unified Python framework implementing many KV cache compression, retrieval, merging, and quantization methods for long… | 65 | 1384 | active |
| yanx27/Pointnet_Pointnet2_pytorch A pure PyTorch implementation of the PointNet and PointNet++ deep learning architectures for point cloud processing. It includes training a… | 32 | 4958 | maintenance |
| hustvl/VAD VAD is an end-to-end autonomous driving framework that models the driving scene as a fully vectorized representation of agents and map elem… | 57 | 1383 | active |
| autonomousvision/unimatch UniMatch is a PyTorch research library implementing a unified transformer-based model for optical flow, stereo matching, and depth estimati… | 32 | 1383 | stable |
| OpenGVLab/DragGAN An unofficial full-featured Python implementation of DragGAN, the interactive point-based image manipulation method built on StyleGAN gener… | 30 | 4944 | maintenance |
| AlmondGod/tinyworlds A minimal Python implementation of DeepMind's Genie autoregressive world model, including a video tokenizer, action tokenizer, and dynamics… | 51 | 1380 | active |