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 |
|---|---|---|---|
| QwenAudio/ThinkSound ThinkSound is a PyTorch implementation of a NeurIPS 2025 framework that generates and edits audio from video, text, or audio inputs using C… | 49 | 1380 | active |
| hudson-and-thames/mlfinlab MlFinLab is a Python library implementing financial machine learning techniques from data structure generation through labeling, feature en… | 32 | 4941 | maintenance |
| fukuball/jieba-php A PHP library for Chinese word segmentation (jieba), ported from the Python jieba project and now independently maintained. It supports pre… | 45 | 1379 | active |
| nomadkaraoke/python-audio-separator A Python package and CLI that separates audio files into stems (vocals, instrumental, drums, bass, etc.) using pre-trained models from Ulti… | 90 | 1378 | active |
| Neuromancer NeuroMANCER is a PyTorch-based differentiable programming library for solving parametric constrained optimization problems, physics-informe… | 69 | 1377 | active |
| giacomelli/GeneticSharp GeneticSharp is a fast, extensible, multithreading C# genetic algorithm library for .NET 6, .NET Standard, and .NET Framework applications.… | 42 | 1376 | stable |
| siyuanliii/masa Official PyTorch implementation of MASA (CVPR 2024 Highlight), a universal instance appearance model that learns to match any objects acros… | 31 | 1376 | active |
| minimaxir/textgenrnn A Python 3 library built on Keras/TensorFlow for easily training char-rnn style neural networks that generate text from any dataset in a fe… | 23 | 4918 | maintenance |
| NGT-labs/NGT NGT is a C++ library and command-line tool from Yahoo Japan for high-speed approximate nearest neighbor search on high-dimensional vector d… | 97 | 1374 | active |
| MACE MACE is a Python library implementing fast and accurate machine learning interatomic potentials using higher-order equivariant message pass… | 86 | 1373 | active |
| andimarafioti/faster-qwen3-tts A Python library for real-time text-to-speech inference with Qwen3-TTS using manual CUDA graph capture, requiring no Flash Attention, vLLM,… | 82 | 1372 | active |
| macanv/BERT-BiLSTM-CRF-NER A TensorFlow implementation of named entity recognition that fine-tunes Google BERT with a BiLSTM-CRF model, primarily targeting Chinese te… | 32 | 4907 | maintenance |
| OpenPrompt OpenPrompt is a PyTorch-based open-source framework for prompt-learning, providing a standard, flexible pipeline of templates and verbalize… | 23 | 4903 | maintenance |
| OpenRaiser/NanoResearch NanoResearch is an autonomous AI research assistant that runs an end-to-end pipeline from a research idea to a complete LaTeX paper. It exe… | 57 | 1369 | active |
| k2-fsa/k2 k2 is a C++/CUDA library with Python bindings that implements differentiable Finite State Automaton (FSA) and Finite State Transducer (FST)… | 60 | 1368 | active |
| OpenPPL/ppl.nn PPLNN is a high-performance deep-learning inference engine written in C++ that runs ONNX models on x86 CPUs and NVIDIA GPUs, with a dedicat… | 32 | 1368 | active |
| mega-sam/mega-sam MegaSaM is a research codebase implementing a deep visual SLAM system that estimates camera parameters and consistent depth maps from casua… | 45 | 1367 | active |
| google/visualblocks Visual Blocks for ML is a Google visual programming framework that provides a no-code node graph editor for building ML pipelines from drag… | 67 | 1366 | active |
| kengz/SLM-Lab SLM Lab is a modular deep reinforcement learning framework built in PyTorch, offering ready-to-use algorithms like PPO, SAC, DQN, and A2C v… | 91 | 1365 | active |
| alibaba/rtp-llm RTP-LLM is Alibaba's high-performance LLM inference engine written in C++/CUDA, optimized with kernels like PagedAttention and FlashAttenti… | 65 | 1364 | active |
| Sense-X/Co-DETR Co-DETR is a PyTorch implementation of DETRs with Collaborative Hybrid Assignments Training, an ICCV 2023 object detection and instance seg… | 32 | 1364 | stable |
| huggingface/finetrainers finetrainers is a Hugging Face library for scalable, memory-optimized training (fine-tuning) of diffusion models, including LoRA training o… | 68 | 1362 | active |
| CMA-ES/pycma pycma is a Python implementation of the CMA-ES (Covariance Matrix Adaptation Evolution Strategy) derivative-free optimization algorithm, al… | 91 | 1361 | stable |
| natasha/natasha Natasha is a Python library that solves basic NLP tasks for the Russian language, including tokenization, sentence segmentation, morphology… | 77 | 1361 | active |
| Swift Transformers A Swift Package providing a transformers-like API for Swift apps, including fast tokenization, chat templating, and reliable model download… | 97 | 1360 | active |
| shivammehta25/Matcha-TTS Matcha-TTS is a PyTorch-based text-to-speech system that uses conditional flow matching for fast, non-autoregressive speech synthesis. It s… | 60 | 1360 | active |
| jhc13/taggui TagGUI is a cross-platform desktop application for quickly adding and editing image tags and captions, aimed at creators of image datasets … | 47 | 1357 | active |
| blei-lab/edward Edward is a Python library for probabilistic modeling, inference, and criticism built on TensorFlow. It supports deep generative models, va… | 23 | 4843 | maintenance |
| uxlfoundation/scikit-learn-intelex Intel's Extension for Scikit-learn is a free AI accelerator that speeds up existing scikit-learn workflows on CPUs and GPUs, claiming up to… | 87 | 1356 | active |
| ZJU-REAL/ClawGUI ClawGUI is a unified Python framework for GUI agents covering the full lifecycle: online reinforcement learning training (ClawGUI-RL with G… | 64 | 1356 | active |
| mmorise/World WORLD is a C++ library for high-quality speech analysis, manipulation, and synthesis based on a vocoder design. It estimates F0 (via DIO/Ha… | 75 | 1355 | stable |
| gempy-project/gempy GemPy is an open-source Python library for implicit 3D structural geological modeling, generating complex models of layers, faults, folds, … | 99 | 1354 | active |
| ClownsharkBatwing/RES4LYF RES4LYF is a ComfyUI custom node collection providing advanced diffusion samplers (RES samplers) that achieve high-quality image generation… | 68 | 1352 | active |
| sjvasquez/handwriting-synthesis A Python implementation of Alex Graves' handwriting synthesis experiments using recurrent neural networks, generating realistic handwritten… | 32 | 4824 | maintenance |
| owlbarn/owl Owl is an OCaml library for scientific and engineering computing, providing n-dimensional arrays, linear algebra, statistics, optimization,… | 62 | 1351 | active |
| bytedance/Lance Lance is a 3B-parameter native unified multimodal model from ByteDance for image and video understanding, generation, and editing, trained … | 52 | 1350 | active |
| christophschuhmann/improved-aesthetic-predictor A CLIP+MLP neural network that predicts how much people on average like an image, trained on AVA dataset ratings. It is widely used for fil… | 32 | 1349 | stable |
| MegEngine/MegEngine MegEngine is a fast, scalable deep learning framework with automatic differentiation, developed in C++ with Python bindings. It unifies tra… | 23 | 4809 | maintenance |
| airbnb/aerosolve Aerosolve is a machine learning library from Airbnb built for human-friendly, interpretable modeling on the JVM. It provides a thrift-based… | 42 | 4807 | maintenance |
| FIND3 FIND (Framework for Internal Navigation and Discovery) is a self-hosted server that determines indoor position using WiFi fingerprints from… | 23 | 4806 | maintenance |
| segment-any-text/wtpsplit wtpsplit is a Python toolkit for segmenting text into sentences or other semantic units using the SaT and WtP deep learning models. It prov… | 93 | 1347 | active |
| xtreme1-io/xtreme1 Xtreme1 is an open-source, self-hosted data labeling and annotation platform for multimodal training data, supporting images, 3D LiDAR poin… | 97 | 1345 | active |
| ace-agent/ace ACE is a Python framework for Agentic Context Engineering that lets language models self-improve by treating contexts as evolving playbooks… | 58 | 1344 | active |
| hao-ai-lab/LookaheadDecoding A Python library implementing Lookahead Decoding, an exact parallel decoding algorithm that accelerates LLM inference without a draft model… | 28 | 1343 | active |
| inference-labs-inc/dsperse DSperse is a proving-system-agnostic tool for verifiable AI that decomposes ONNX neural network models into circuit-compatible segments and… | 76 | 1342 | active |
| XTXMarkets/ternfs TernFS is an exabyte-scale, multi-region distributed file system written in C++, designed for storing large immutable files on commodity ha… | 60 | 1342 | active |
| rstudio/tensorflow An R package that provides full access to the TensorFlow API from R via reticulate, bridging R users to TensorFlow's Python implementation.… | 56 | 1340 | active |
| alibaba/graph-learn Graph-Learn (formerly AliGraph) is a distributed framework for developing and applying large-scale graph neural networks, with a training l… | 32 | 1340 | active |
| open-edge-platform/geti Geti is an open-source, end-to-end Vision AI application from Intel that takes users from raw images to deployed computer vision models, ru… | 92 | 1338 | active |
| jonathan-laurent/AlphaZero.jl A generic, simple, and fast Julia implementation of DeepMind's AlphaZero algorithm for training game-playing agents via self-play and MCTS.… | 69 | 1338 | active |
| BytedTsinghua-SIA/CUDA-Agent CUDA-Agent is a large-scale agentic reinforcement learning system from ByteDance Seed and Tsinghua that trains LLMs to generate high-perfor… | 53 | 1338 | active |
| PKU-YuanGroup/MagicTime MagicTime is a metamorphic time-lapse video generation pipeline built on diffusion-based text-to-video models, with a MagicAdapter, dynamic… | 56 | 1337 | active |
| CarperAI/trlx trlX is a distributed training framework for fine-tuning large language models with reinforcement learning from human feedback (RLHF), supp… | 23 | 4756 | maintenance |
| awslabs/dgl-ke DGL-KE is a high-performance Python package built on Deep Graph Library (DGL) for training, evaluating, and inferring knowledge graph embed… | 60 | 1335 | active |
| pykalman/pykalman pykalman is a Python library implementing the Kalman Filter, Kalman Smoother, Unscented Kalman Filter, and EM algorithm for state estimatio… | 68 | 1334 | active |
| IrisRainbowNeko/genshin_auto_fish A Genshin Impact auto-fishing AI built from a YOLOX object detection model (fish and rod landing point localization) and a DQN reinforcemen… | 23 | 4748 | maintenance |
| mapillary/inplace_abn A PyTorch extension library implementing In-Place Activated BatchNorm (InPlace-ABN), which redefines BN plus nonlinear activation as a sing… | 67 | 1333 | stable |
| braindecode/braindecode Braindecode is an open-source Python toolbox built on PyTorch for decoding raw electrophysiological brain signals such as EEG, ECoG, and ME… | 97 | 1332 | active |
| facebookincubator/AITemplate AITemplate is a Python framework that compiles deep neural networks into high-performance CUDA (NVIDIA) or HIP (AMD) C++ code for fast fp16… | 67 | 4730 | maintenance |
| Vahe1994/AQLM Official PyTorch implementation of AQLM, an extreme LLM compression method via additive quantization, extended with PV-Tuning for finetunin… | 54 | 1329 | active |
| PrunaAI/pruna Pruna is an open-source Python model optimization framework that makes AI models faster, smaller, cheaper, and greener via caching, quantiz… | 82 | 1326 | active |
| Tongjilibo/bert4torch bert4torch is a PyTorch library providing an elegant reimplementation of transformer models (BERT, RoBERTa, T5, GPT, ChatGLM, LLaMA, etc.) … | 76 | 1326 | active |
| jishengpeng/WavTokenizer WavTokenizer is a state-of-the-art discrete neural audio codec that compresses speech, music, and general audio into only 40 or 75 discrete… | 23 | 1325 | active |
| inference-labs-inc/JSTprove JSTprove is a Rust CLI toolkit that generates zero-knowledge proofs of machine learning inference on ONNX models, built on Polyhedra Networ… | 64 | 1324 | active |
| tensorflow/lucid Lucid is a collection of infrastructure and tools for research in neural network interpretability, built on TensorFlow 1.x. It provides fea… | 10 | 4704 | maintenance |
| DBraun/DawDreamer DawDreamer is a Python framework that provides digital audio workstation features, letting users compose graphs of audio processors includi… | 84 | 1322 | active |
| yeyupiaoling/VoiceprintRecognition-Pytorch A PyTorch-based voiceprint recognition (speaker recognition) framework implementing models such as ECAPA-TDNN, ResNetSE, ERes2Net, and CAM+… | 75 | 1322 | active |
| sebp/scikit-survival scikit-survival is a Python library for survival analysis (time-to-event modeling) built on top of scikit-learn. It provides survival model… | 84 | 1321 | active |
| stared/livelossplot A Python library that draws live training loss and metric plots inside Jupyter Notebooks for Keras, PyTorch, and other deep learning framew… | 85 | 1320 | stable |
| sanchit-gandhi/whisper-jax An optimized JAX implementation of OpenAI's Whisper speech recognition model, built on Hugging Face Transformers, offering up to 70x faster… | 31 | 4680 | maintenance |
| ChenmienTan/RL2 RL2 (Ray Less Reinforcement Learning) is a concise Python library for post-training large language models with reinforcement learning, SFT,… | 54 | 1316 | active |
| sicara/easy-few-shot-learning A Python library (easyfsl) with ready-to-use code and tutorial notebooks for few-shot image classification and meta-learning, built on PyTo… | 23 | 1316 | stable |
| General Navigation Models Official code and pre-trained checkpoints for the GNM, ViNT, and NoMaD family of general-purpose goal-conditioned visual navigation policie… | 20 | 1315 | active |
| arkflow-rs/arkflow ArkFlow is a high-performance stream processing engine written in Rust on top of Tokio, connecting configurable inputs (Kafka, MQTT, HTTP, … | 88 | 1314 | active |
| Alibaba-NLP/ZeroSearch ZeroSearch is a reinforcement learning framework from Alibaba's Tongyi Lab that trains LLMs to use search by simulating search engine resul… | 32 | 1313 | active |
| OpenTeleVision/TeleVision Open-TeleVision is an open-source immersive robot teleoperation system that streams stereoscopic visual feedback to VR headsets (Apple Visi… | 25 | 1313 | active |
| ai-dynamo/nixl NVIDIA Inference Xfer Library (NIXL) is a C++/Python library that accelerates point-to-point communication in AI inference frameworks like … | 87 | 1312 | active |
| vye16/shape-of-motion Shape of Motion is a Python research codebase for 4D reconstruction of dynamic scenes from a single monocular video, based on the ICCV 2025… | 27 | 1311 | active |
| Hopsworks Hopsworks is an open-source, data-intensive AI platform (an 'AI Lakehouse') built around a Python-centric Feature Store with online/offline… | 23 | 1310 | active |
| StarlightSearch/EmbedAnything EmbedAnything is a high-performance, memory-safe embedding pipeline written in Rust (with Python bindings) that generates embeddings from t… | 86 | 1309 | active |
| tjake/Jlama Jlama is a modern LLM inference engine written in Java, supporting popular transformer models like Llama, Mistral, Gemma, and Qwen with qua… | 48 | 1309 | active |
| Parskatt/RoMa RoMa (romatch) is a Python library for robust dense feature matching between image pairs, estimating pixel-dense warps and reliable certain… | 46 | 1309 | active |
| donydchen/mvsplat MVSplat is a PyTorch implementation of an ECCV 2024 Oral model that predicts 3D Gaussians from sparse multi-view images in a single feed-fo… | 58 | 1308 | active |
| openai/simple-evals A lightweight Python library from OpenAI for evaluating language models against benchmarks like MMLU, GPQA, MATH, HumanEval, SimpleQA, Heal… | 57 | 4638 | maintenance |
| SakanaAI/text-to-lora Text-to-LoRA (T2L) is a hypernetwork that generates LoRA adapters for large language models in a single forward pass, using only a natural … | 28 | 1307 | active |
| albermax/innvestigate iNNvestigate is a Python toolbox providing a common interface and out-of-the-box implementations of many neural network explanation methods… | 26 | 1307 | active |
| huawei-noah/Efficient-Computing A collection of efficient deep learning methods from Huawei Noah's Ark Lab, covering model compression, knowledge distillation, pruning, qu… | 32 | 1306 | active |
| turboderp-org/exllamav2 ExLlamaV2 is a fast Python inference library for running large language models locally on modern consumer GPUs, with support for EXL2/GPTQ … | 58 | 4631 | maintenance |
| derrian-distro/LoRA_Easy_Training_Scripts A PySide6 desktop GUI that wraps Kohya's sd-scripts to simplify training LoRA, LoCon, and other LoRA-type models for Stable Diffusion. It s… | 30 | 1304 | active |
| bambinos/bambi Bambi is a high-level Bayesian model-building interface for Python built on top of the PyMC probabilistic programming framework. It uses a … | 93 | 1303 | active |
| yizhongw/self-instruct Self-Instruct is a framework and research codebase for aligning pretrained language models with instructions using data generated by the mo… | 32 | 4612 | maintenance |
| buoyancy99/diffusion-forcing Official research code for the NeurIPS paper 'Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion', implementing a metho… | 61 | 1299 | active |
| STVIR/pysot PySOT is a Python research platform by SenseTime for single object visual tracking, implementing algorithms such as SiamRPN, SiamRPN++, DaS… | 41 | 4602 | maintenance |
| jbarrow/commonforms CommonForms is a Python package and CLI that uses trained object-detection models (FFDNet-S/L) to automatically detect form fields in a PDF… | 59 | 1297 | active |
| thu-ml/Motus Motus is the official implementation of a unified latent action world model for robotics, combining a video generation model, a vision-lang… | 40 | 1297 | active |
| PantoMatrix/PantoMatrix PantoMatrix is an open-source research project that generates 3D face and body animation from speech audio, including the EMAGE model for h… | 32 | 1297 | active |
| ZhengyiLuo/PHC Official implementation of the ICCV 2023 paper 'Perpetual Humanoid Control for Real-time Simulated Avatars'. It provides a Python codebase … | 41 | 1296 | active |
| endee-io/endee Endee is a high-performance C++ vector database designed to handle up to 1B vectors on a single node, supporting dense, sparse (BM25), and … | 10 | 1296 | active |
| ictnlp/StreamSpeech StreamSpeech is an 'All in One' seamless model for offline and simultaneous speech recognition, speech translation, and speech synthesis, p… | 34 | 1294 | active |
| zju3dv/MatchAnything MatchAnything is a deep learning model for universal cross-modality image matching, released as research code accompanying a TPAMI 2026 pap… | 64 | 1293 | active |