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 |
|---|---|---|---|
| nutonomy/nuscenes-devkit The official Python devkit for the nuScenes dataset, a large-scale autonomous driving dataset from Motional. It provides dataset loading, v… | 62 | 2807 | stable |
| satijalab/seurat Seurat is an R toolkit for single-cell genomics developed by the Satija Lab, providing a complete pipeline for analyzing single-cell RNA-se… | 88 | 2803 | stable |
| mll-lab-nu/RAGEN RAGEN is a Python framework for training reasoning LLM agents with multi-turn reinforcement learning using the StarPO algorithm. It also pr… | 62 | 2803 | active |
| pytorch/xla PyTorch/XLA is a Python package that connects the PyTorch deep learning framework to XLA devices such as Google Cloud TPUs via the XLA deep… | 63 | 2801 | active |
| rom1504/clip-retrieval A Python toolkit for computing CLIP embeddings for images and text and building a semantic search/retrieval system on top of them. It inclu… | 52 | 2797 | active |
| Mega4alik/ollm oLLM is a lightweight Python library for large-context LLM inference built on Hugging Face Transformers and PyTorch. It offloads weights an… | 57 | 2795 | active |
| lucidrains/DALLE2-pytorch A PyTorch implementation of OpenAI's DALL-E 2 text-to-image synthesis model, focusing on the diffusion prior network that predicts image em… | 23 | 11302 | maintenance |
| rhasspy/piper Piper is a fast, local neural text-to-speech system that runs offline on modest hardware, including Raspberry Pi devices. It offers many pr… | 10 | 11290 | maintenance |
| Autodistill Autodistill is a Python library that uses large foundation vision models (like Grounding DINO, Grounded SAM, and CLIP) to automatically lab… | 64 | 2784 | active |
| FasterDecoding/Medusa Medusa is a framework that accelerates LLM text generation by adding multiple decoding heads to an existing model, avoiding the need for a … | 19 | 2776 | active |
| Stan Stan is a C++ probabilistic programming library for full Bayesian inference via NUTS/Hamiltonian Monte Carlo, approximate inference via ADV… | 88 | 2774 | stable |
| pyro-ppl/numpyro NumPyro is a lightweight probabilistic programming library providing a NumPy backend for Pyro, built on JAX for automatic differentiation a… | 94 | 2770 | active |
| google-research/tabfm TabFM is a scikit-learn compatible tabular foundation model from Google Research that performs zero-shot classification and regression on t… | 63 | 2769 | active |
| apple/turicreate Turi Create is a Python library from Apple that simplifies building custom machine learning models for tasks like image classification, obj… | 10 | 11155 | maintenance |
| voxelmorph/voxelmorph VoxelMorph is a Python library for learning-based image registration and alignment, using unsupervised deep learning to model deformations … | 75 | 2754 | active |
| thewh1teagle/kokoro-onnx A Python library that runs the Kokoro text-to-speech model via ONNX Runtime, supporting CPU and GPU inference. It provides multi-language T… | 65 | 2750 | active |
| lightseekorg/tokenspeed TokenSpeed is a high-performance LLM inference engine designed for agentic workloads, aiming for TensorRT-LLM-level performance with vLLM-l… | 78 | 2743 | active |
| TWIST GMR is a Python library for general motion retargeting that converts human motions into joint commands for diverse humanoid robots in real … | 49 | 2741 | active |
| artidoro/qlora QLoRA is the official implementation of the QLoRA paper, an efficient finetuning approach that backpropagates through a frozen 4-bit quanti… | 30 | 11035 | maintenance |
| huggingface/OpenEnv OpenEnv is a Hugging Face framework for creating, deploying, and consuming isolated execution environments used in agentic reinforcement le… | 85 | 2728 | active |
| qualcomm/aimet AIMET (AI Model Efficiency Toolkit) is a Python library from Qualcomm providing advanced quantization and compression techniques for traine… | 98 | 2723 | active |
| intel/neural-compressor Intel Neural Compressor is an open-source Python library providing state-of-the-art low-bit quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4),… | 95 | 2723 | active |
| prophesier/diff-svc Diff-SVC is a deep learning project that performs singing voice conversion using diffusion models, transforming input singing audio into a … | 58 | 2719 | active |
| yuqinie98/PatchTST Official PyTorch implementation of PatchTST, an ICLR 2023 Transformer model for long-term time series forecasting based on patching and cha… | 32 | 2719 | stable |
| mljs/ml ml.js is an umbrella library bundling the mljs organization's machine learning tools for JavaScript, including clustering, classification, … | 23 | 2719 | active |
| secretflow/secretflow SecretFlow is a unified Python framework for privacy-preserving data analysis and machine learning. It layers cryptographic devices (MPC, H… | 62 | 2713 | active |
| openai/privacy-filter OpenAI Privacy Filter is a small bidirectional token-classification model (1.5B total, 50M active parameters) for detecting and masking per… | 46 | 2711 | active |
| yuweihao/MambaOut MambaOut is a PyTorch implementation of Gated CNN models from the CVPR 2025 paper 'MambaOut: Do We Really Need Mamba for Vision?', which qu… | 16 | 2710 | stable |
| KimMeen/Time-LLM Time-LLM is the official PyTorch implementation of an ICLR 2024 paper that reprograms frozen large language models (Llama, GPT-2, BERT) for… | 44 | 2707 | active |
| dreamzero0/dreamzero DreamZero is NVIDIA's World Action Model (WAM) that jointly predicts future video and actions from a pretrained video diffusion backbone, e… | 48 | 2700 | active |
| ZHZisZZ/dllm dLLM is a Python library that unifies training, inference, and evaluation of diffusion language models such as LLaDA and Dream. It builds o… | 56 | 2699 | active |
| huggingface/text-generation-inference Text Generation Inference (TGI) is a Rust, Python and gRPC toolkit for deploying and serving large language models with high performance, p… | 10 | 10877 | maintenance |
| roboflow/maestro maestro is a Python library from Roboflow that streamlines fine-tuning of multimodal vision-language models such as Florence-2, PaliGemma 2… | 61 | 2696 | active |
| naiveHobo/InvoiceNet InvoiceNet is a deep neural network application with a GUI for extracting structured information from invoice documents in PDF, JPG, and PN… | 32 | 2696 | active |
| tryolabs/norfair Norfair is a lightweight, customizable Python library for real-time multi-object tracking that works with any detector outputting (x, y) co… | 28 | 2689 | stable |
| JIA-Lab-research/LISA LISA (Large Language Instructed Segmentation Assistant) is a multimodal large language model that performs reasoning-based image segmentati… | 29 | 2687 | active |
| openai/DALL-E The official PyTorch package for the discrete VAE (dVAE) component of OpenAI's DALL·E model. It does not include the transformer that gener… | 10 | 10820 | maintenance |
| DDSP-SVC DDSP-SVC is an open-source singing voice conversion system built on Differentiable Digital Signal Processing, designed as a free AI voice c… | 65 | 2682 | active |
| google-deepmind/mctx Mctx is a JAX-native Python library implementing Monte Carlo tree search algorithms such as AlphaZero, MuZero, and Gumbel MuZero. It suppor… | 81 | 2678 | active |
| stochasticai/xTuring xTuring is a Python library for fine-tuning, evaluating, and running open-source large language models such as LLaMA, GPT-J, GPT-2, Qwen, a… | 65 | 2675 | active |
| georgia-tech-db/evadb EvaDB is a Python database system for building AI-powered applications, exposing a SQL API that queries structured data, videos, and unstru… | 10 | 2674 | active |
| AMAP-ML/SkillClaw SkillClaw is a Python-based framework that lets AI agent skills evolve collectively from real interactions across sessions, agents, devices… | 55 | 2669 | active |
| open-gigaai/giga-brain-0 GigaBrain-0/0.7 is an open-source vision-language-action (VLA) model family for generalist embodied agents, powered by world models and a t… | 60 | 2659 | active |
| Const-me/Whisper A Windows port of whisper.cpp that runs OpenAI's Whisper speech recognition model on the GPU via DirectCompute (Direct3D 11 compute shaders… | 66 | 10678 | maintenance |
| SeldonIO/alibi Alibi is a Python library providing algorithms for explaining and interpreting machine learning models, including black-box, white-box, loc… | 40 | 2646 | active |
| plexe-ai/plexe Plexe is a Python library and CLI that builds machine learning models from natural language descriptions using a multi-agent AI workflow. Y… | 56 | 2645 | active |
| facebookresearch/ParlAI ParlAI is a Python framework from Facebook AI Research for sharing, training, and evaluating dialogue models across many openly available d… | 10 | 10618 | maintenance |
| ultralytics/yolov3 Ultralytics' PyTorch implementation of YOLOv3, YOLOv3-SPP, and YOLOv3-tiny for real-time object detection. It provides training, validation… | 66 | 10613 | maintenance |
| AudioLDM AudioLDM 2 is a Python library and CLI for generating audio, music, and speech from text prompts using latent diffusion models. It includes… | 29 | 2642 | active |
| EMI-Group/evox EvoX is a distributed GPU-accelerated framework for evolutionary computation, compatible with PyTorch and built on JAX. It provides 50+ evo… | 97 | 2640 | active |
| AudioMuse-AI AudioMuse-AI is a self-hosted application that performs sonic analysis on your music library to generate intelligent, AI-driven playlists w… | 80 | 2629 | active |
| OmniSVG/OmniSVG OmniSVG is a family of end-to-end multimodal SVG generation models built on pre-trained Vision-Language Models, released with inference cod… | 48 | 2626 | active |
| lucidrains/audiolm-pytorch A PyTorch implementation of AudioLM, Google Research's language modeling approach to audio generation, including a MIT-licensed SoundStream… | 34 | 2625 | active |
| meta-pytorch/torchrec TorchRec is a PyTorch domain library for building recommendation systems at scale. It provides distributed sharding of large embedding tabl… | 95 | 2616 | active |
| google-coral/coralnpu Coral NPU is an open-source neural processing unit (NPU) hardware IP core from Google Research, built on the 32-bit RISC-V ISA with matrix,… | 72 | 2604 | active |
| apache/hamilton Apache Hamilton is a lightweight Python library for defining directed acyclic graphs (DAGs) of data transformations as plain, testable Pyth… | 89 | 2603 | active |
| kayba-ai/agentic-context-engine Agentic Context Engine (ACE) is a Python library that lets LLM-based agents learn from experience by maintaining and evolving an evolving c… | 82 | 2602 | active |
| crowsonkb/k-diffusion A PyTorch library implementing Karras et al. (2022) diffusion models with enhancements like improved sampling algorithms and transformer-ba… | 49 | 2599 | active |
| atong01/conditional-flow-matching TorchCFM is a PyTorch library implementing Conditional Flow Matching (CFM), a simulation-free training objective for continuous normalizing… | 72 | 2596 | active |
| asteroid-team/asteroid Asteroid is a PyTorch-based audio source separation toolkit for researchers, providing modular building blocks (filterbanks, encoders, mask… | 66 | 2593 | active |
| median-research-group/LibMTL LibMTL is an open-source PyTorch library for Multi-Task Learning (MTL). It provides implementations of many MTL architectures and gradient-… | 38 | 2593 | active |
| facebookresearch/demucs Demucs is a state-of-the-art music source separation model from Meta AI that splits songs into stems like drums, bass, and vocals using a h… | 10 | 10373 | maintenance |
| Google Colaboratory The official Python libraries shipped inside Google Colaboratory, Google's free hosted Jupyter notebook environment for machine learning ed… | 77 | 2591 | active |
| kijai/ComfyUI-HunyuanVideoWrapper A set of custom ComfyUI nodes that wrap Tencent's HunyuanVideo text-to-video and image-to-video diffusion model for use inside ComfyUI work… | 35 | 2591 | active |
| KomputeProject/kompute Kompute is a general-purpose GPU compute framework built on Vulkan that works across vendor GPUs (AMD, NVIDIA, Qualcomm, etc.) with both C+… | 63 | 2577 | active |
| scverse/scanpy Scanpy is a scalable Python toolkit for analyzing single-cell gene expression data, built alongside the anndata data structure. It provides… | 91 | 2574 | stable |
| sarah-quinones/faer-rs faer is a pure Rust linear algebra library providing low-level routines and a high-level ergonomic wrapper for dense matrix computations. I… | 80 | 2573 | active |
| pykeio/ort ort is a safe, fast Rust binding for Microsoft's ONNX Runtime, providing hardware-accelerated machine learning inference and training for O… | 91 | 2569 | active |
| huggingface/lighteval Lighteval is Hugging Face's all-in-one toolkit for evaluating large language models across multiple inference backends (Transformers, vLLM,… | 72 | 2563 | active |
| ymcui/Chinese-BERT-wwm A collection of Chinese pre-trained language models (BERT-wwm, BERT-wwm-ext, RoBERTa-wwm-ext, RBT3, etc.) built with Whole Word Masking, re… | 63 | 10231 | maintenance |
| Hugging Face JavaScript SDK A collection of TypeScript JavaScript libraries for interacting with the Hugging Face ecosystem, including Hub repository management, unifi… | 95 | 2560 | active |
| LTH14/JiT A PyTorch/GPU re-implementation of JiT (Just image Transformer), a minimalist pixel-space diffusion model for high-resolution image generat… | 39 | 2553 | active |
| DeepDetect DeepDetect is an open-source deep learning runtime, CLI, and REST server written in C++ for training and inference across images, text, tab… | 95 | 2551 | active |
| mozilla/TTS A deep learning library for advanced text-to-speech generation, built on PyTorch with models like Tacotron2, Glow-TTS, and various vocoders… | 23 | 10173 | maintenance |
| SeldonIO/alibi-detect Alibi Detect is a Python library providing algorithms for outlier, adversarial, and drift detection across tabular data, text, images, and … | 49 | 2548 | active |
| data-infra/cube-studio CubeStudio is an open-source, cloud-native, all-in-one AI platform covering the full machine learning lifecycle (MLOps/MaaS/LLMOps), includ… | 74 | 2546 | active |
| yifan123/flow_grpo Flow-GRPO is the official PyTorch implementation of a NeurIPS 2025 paper that trains flow matching models (e.g., SD3.5, FLUX.1, Qwen-Image,… | 52 | 2543 | active |
| SafeAILab/EAGLE EAGLE is the official implementation of the EAGLE family of speculative decoding algorithms (EAGLE-1/2/3) for accelerating LLM text generat… | 54 | 2542 | active |
| learning-at-home/hivemind Hivemind is a PyTorch library for decentralized deep learning across the Internet, enabling training of large models on hundreds of volunte… | 53 | 2541 | active |
| chatdoc-com/OCRFlux OCRFlux is a Python toolkit built on a 3B-parameter vision-language model that converts PDFs and images into clean Markdown, handling compl… | 50 | 2532 | active |
| LyCORIS LyCORIS is a Python library implementing parameter-efficient fine-tuning algorithms (LoRA/LoCon, LoHa, LoKr, IA3, DyLoRA, and more) for Sta… | 99 | 2530 | active |
| facebookresearch/nougat Nougat is Meta's neural OCR model that parses academic PDFs into structured Markdown, understanding LaTeX math and tables. It ships as a Py… | 20 | 10082 | maintenance |
| rpautrat/SuperPoint A TensorFlow (with PyTorch conversion) implementation of the SuperPoint self-supervised interest point detector and descriptor network. It … | 37 | 2528 | stable |
| mflux-community/mflux MFLUX is a native MLX implementation of state-of-the-art generative image and video models (Flux, Qwen-Image, Z-Image, and others), ported … | 91 | 2526 | active |
| Farama-Foundation/Minigrid Minigrid is a Python library of lightweight, easily configurable discrete grid-world environments for reinforcement learning research, foll… | 80 | 2523 | active |
| bilibili/Index-anisora Index-AniSora is Bilibili's open-source anime video generation model, capable of creating video shots in diverse anime styles from images, … | 60 | 2523 | active |
| amap-cvlab/ABot-World ABot-World is an open-source interactive world model that generates infinite, explorable video worlds from a single image, running on a sin… | 57 | 2513 | active |
| oegedijk/explainerdashboard A Python library for quickly building interactive Explainable AI dashboards that explain the inner workings of scikit-learn compatible mach… | 68 | 2512 | active |
| youssofal/MTPLX MTPLX is a native macOS app and CLI that runs local LLMs on Apple Silicon roughly twice as fast using native multi-token prediction (MTP) s… | 80 | 2511 | active |
| HiDream-ai/HiDream-I1 HiDream-I1 is an open-source 17B-parameter text-to-image generative foundation model based on a Sparse Diffusion Transformer, with full and… | 31 | 2510 | active |
| google-deepmind/sonnet Sonnet is a Python library built on TensorFlow 2 that provides simple, composable abstractions (snt.Module) for constructing neural network… | 66 | 9973 | maintenance |
| scikit-learn-contrib/category_encoders A scikit-learn-contrib library providing a collection of sklearn-compatible transformers for encoding categorical variables into numeric fo… | 91 | 2506 | active |
| yformer/EfficientSAM EfficientSAM is an efficient image segmentation model that leverages masked image pretraining to provide a lightweight alternative to Meta'… | 27 | 2498 | active |
| twistedfall/opencv-rust Rust bindings for the OpenCV computer vision library, generated automatically via Clang. It exposes OpenCV 3.4 (deprecated), 4.x, and 5.x A… | 75 | 2494 | active |
| sthalles/SimCLR A PyTorch reference implementation of SimCLR, a self-supervised contrastive learning framework for learning visual representations from unl… | 23 | 2494 | stable |
| google/tunix Tunix is a lightweight JAX-based library for post-training large language models, supporting supervised fine-tuning, preference optimizatio… | 81 | 2491 | active |
| ipazc/mtcnn A Python library implementing the MTCNN (Multitask Cascaded Convolutional Networks) algorithm for face detection and facial landmark alignm… | 23 | 2490 | stable |
| ppogg/YOLOv5-Lite YOLOv5-Lite is a lightweight object detection model family evolved from YOLOv5, with models as small as ~900KB (int8) that run 10-15+ FPS o… | 23 | 2488 | active |
| thuml/Autoformer Autoformer is the official PyTorch implementation of the NeurIPS 2021 paper 'Autoformer: Decomposition Transformers with Auto-Correlation f… | 32 | 2485 | stable |
| homuler/MediaPipeUnityPlugin A Unity native plugin that ports the MediaPipe C++ API to C#, enabling MediaPipe graphs and solutions to run inside Unity applications. It … | 56 | 2484 | active |