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 |
|---|---|---|---|
| FluidAudio A Swift SDK providing fully local, low-latency audio AI on Apple devices, including speech-to-text (Parakeet), speaker diarization (Pyannot… | 82 | 3139 | active |
| timescale/pgvectorscale pgvectorscale is a PostgreSQL extension written in Rust that complements pgvector with a StreamingDiskANN index, statistical binary quantiz… | 87 | 3136 | active |
| docarray/docarray DocArray is a Python library for representing, transmitting, storing, and retrieving multimodal data such as text, images, audio, and video… | 10 | 3125 | 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 |
| 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 |
| 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 |
| lammps/lammps LAMMPS (Large-scale Atomic/Molecular Massively Parallel Simulator) is an open-source classical molecular dynamics simulation code focused o… | 95 | 3104 | stable |
| 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 |
| TradeMaster-NTU/TradeMaster TradeMaster is an open-source Python platform for quantitative trading powered by reinforcement learning, covering the full pipeline of alg… | 30 | 3087 | active |
| kubernetes-sigs/kueue Kueue is a Kubernetes-native job queueing system providing APIs and controllers that decide when batch, HPC, and AI/ML workloads should be … | 95 | 3086 | active |
| DamRsn/NeuralNote NeuralNote is an audio plugin (VST3/AU/Standalone) that performs audio-to-MIDI transcription using deep learning, based on Spotify's basic-… | 86 | 3084 | active |
| guillaume-be/rust-bert A Rust-native library providing ready-to-use NLP pipelines and transformer-based models (BERT, DistilBERT, GPT-2, RoBERTa, BART, etc.), por… | 57 | 3084 | active |
| microsoft/MoGe MoGe is a deep learning model from Microsoft Research that recovers 3D geometry from a single open-domain image, predicting metric point ma… | 68 | 3077 | active |
| sonos/tract Tract is Sonos' tiny, self-contained neural-network inference engine written in Rust. It loads ONNX, TensorFlow/TFLite, and NNEF models, op… | 97 | 3075 | active |
| leggedrobotics/rsl_rl RSL-RL is a GPU-accelerated, lightweight reinforcement learning library for robotics research, featuring algorithms like PPO and Student-Te… | 97 | 3075 | active |
| kairos-agi/kairos Kairos is the official open-source implementation of a 4B-parameter native cross-embodiment world model that unifies video understanding, f… | 59 | 3066 | active |
| Neural Amp Modeler Neural Amp Modeler (NAM) is a Python project that trains neural network models to emulate guitar amplifiers and exports them as .nam files.… | 80 | 3060 | active |
| hegelai/prompttools PromptTools is an open-source Python library for testing, experimenting with, and evaluating prompts, LLMs (OpenAI, Anthropic, LLaMA, etc.)… | 46 | 3056 | active |
| UKGovernmentBEIS/inspect_ai Inspect is an open-source Python framework from the UK AI Security Institute for running large language model evaluations, with composable … | 95 | 3054 | active |
| yuchenlin/rebiber Rebiber is a Python CLI tool that replaces unofficial or arXiv BibTeX entries with official records from DBLP or the ACL Anthology, keeping… | 84 | 3037 | active |
| facebookresearch/Pearl Pearl is a production-ready reinforcement learning agent library developed by Meta's Applied Reinforcement Learning team. It provides modul… | 71 | 3033 | active |
| wasserth/TotalSegmentator TotalSegmentator is a Python command-line tool that robustly segments over 100 anatomical structures in CT and MR images using deep learnin… | 66 | 3033 | active |
| deepseek-ai/DreamCraft3D Official PyTorch implementation of DreamCraft3D, an ICLR 2024 hierarchical 3D content generation method that turns a single 2D image into a… | 32 | 3032 | stable |
| tensorflow/agents TF-Agents is a TensorFlow-based Python library for building, training, and deploying reinforcement learning and contextual bandit algorithm… | 57 | 3028 | active |
| luminal-ai/luminal Luminal is a high-performance general-purpose ML inference compiler written in Rust that lowers models to a minimal 15-op dataflow IR and c… | 77 | 3026 | active |
| Eladlev/AutoPrompt AutoPrompt is a Python framework for optimizing LLM prompts using Intent-based Prompt Calibration. It iteratively generates challenging edg… | 39 | 3017 | active |
| osmr/imgclsmob A research sandbox providing (re)implementations of numerous deep learning computer vision models for classification, segmentation, detecti… | 23 | 3012 | active |
| pytorch/TensorRT Torch-TensorRT is a compiler library that accelerates PyTorch model inference on NVIDIA GPUs using TensorRT. It supports just-in-time compi… | 93 | 3009 | active |
| confident-ai/deepteam DeepTeam is an open-source Python framework for red teaming LLM systems, AI agents, RAG pipelines, and chatbots. It simulates adversarial a… | 65 | 3005 | active |
| pytorch/ao TorchAO is a PyTorch-native library for model optimization through quantization and sparsity. It supports quantizing weights, gradients, op… | 87 | 3001 | active |
| signerless/llm-checker A Node.js CLI tool that scans your hardware (GPU, VRAM, CPU, memory) and recommends which LLM or small language models you can run locally,… | 80 | 2998 | active |
| ulab-uiuc/LLMRouter LLMRouter is an open-source Python library that intelligently routes queries to the most suitable LLM based on task complexity, cost, and p… | 62 | 2995 | active |
| mazzzystar/Queryable Queryable is an open-source iOS app that runs Apple's MobileCLIP (formerly OpenAI's CLIP) entirely on-device to search your photo album wit… | 59 | 2993 | active |
| benedekrozemberczki/pytorch_geometric_temporal PyTorch Geometric Temporal is a temporal (dynamic) extension library for PyTorch Geometric providing spatiotemporal signal processing with … | 64 | 2992 | active |
| hosseinmoein/DataFrame A high-performance, templatized C++ library providing a heterogeneous DataFrame container for in-memory data exploration, transformation, a… | 95 | 2990 | active |
| openai/shap-e Shap-E is OpenAI's official release of a generative model that produces 3D implicit functions conditioned on text prompts or images. It inc… | 30 | 12263 | maintenance |
| Biohub/esm A Python library and model release providing ESMC protein language models, ESMFold2 structure prediction, and the ESM Atlas of 6.8 billion … | 88 | 2987 | active |
| sovrasov/flops-counter.pytorch A Python library (ptflops) that computes the theoretical number of multiply-add operations (FLOPs) and parameter counts for neural network … | 36 | 2983 | active |
| bethgelab/foolbox Foolbox is a Python library for generating adversarial examples that fool deep neural networks, with state-of-the-art gradient-based and de… | 44 | 2979 | active |
| patrick-kidger/equinox Equinox is a Python library providing neural networks and scientific computing utilities for JAX, using PyTorch-like class-based syntax whe… | 88 | 2976 | stable |
| ekzhu/datasketch datasketch is a Python library of probabilistic data structures (MinHash, Weighted MinHash, HyperLogLog, HyperLogLog++) and indexes (MinHas… | 90 | 2969 | stable |
| LIME Lime (Local Interpretable Model-agnostic Explanations) is a Python library that explains the predictions of any machine learning classifier… | 23 | 12170 | maintenance |
| michaelfeil/infinity Infinity is a high-throughput, low-latency serving engine that exposes text-embedding, reranking, CLIP, CLAP and ColPali models via an Open… | 96 | 2966 | active |
| 0xemmkty/QuantMuse A Python-based quantitative trading system combining factor analysis, ML models, and LLM-powered market analysis with real-time market data… | 33 | 2964 | active |
| zju3dv/LoFTR LoFTR is a detector-free local image feature matching method using Transformers, released with PyTorch inference and training code plus pre… | 32 | 2963 | stable |
| sunsmarterjie/yolov12 YOLOv12 is a PyTorch implementation of attention-centric real-time object detectors, published at NeurIPS 2025. It provides detection model… | 65 | 2961 | active |
| mit-nlp/MITIE MITIE is a C++ library and toolset for information extraction, providing named entity recognition and binary relation detection built on dl… | 39 | 2960 | active |
| karpathy/char-rnn char-rnn is a Torch/Lua implementation of multi-layer recurrent neural networks (RNN, LSTM, GRU) for character-level language modeling. It … | 32 | 12120 | maintenance |
| li-plus/chatglm.cpp A pure C++ implementation of ChatGLM-6B, ChatGLM2-6B, ChatGLM3, and GLM-4(V) based on ggml, enabling memory-efficient real-time chatting on… | 21 | 2957 | active |
| dreamquark-ai/tabnet A PyTorch implementation of the TabNet deep learning architecture for tabular data, providing attentive, interpretable models for classific… | 23 | 2956 | stable |
| iscyy/ultralyticsPro A PyTorch-based collection of improved YOLO-family object detection models (YOLOv5 through YOLOv13, RT-DETR) with pluggable modules for bac… | 45 | 2954 | active |
| TylerYep/torchinfo torchinfo is a Python library that provides Keras-style model summaries for PyTorch, showing layer types, input/output shapes, parameter co… | 66 | 2953 | active |
| KichangKim/DeepDanbooru DeepDanbooru is a Python/TensorFlow system that estimates Danbooru-style tags for anime-style girl images using multi-label classification.… | 60 | 2948 | active |
| microsoft/table-transformer Table Transformer (TATR) is a deep learning object detection model from Microsoft for detecting, recognizing, and extracting tables from un… | 10 | 2948 | active |
| turboderp/exllama ExLlama is a standalone Python/C++/CUDA implementation of the Llama architecture optimized for running 4-bit GPTQ quantized weights on mode… | 30 | 2947 | active |
| KevinWang676/Bark-Voice-Cloning A one-click hub of Gradio Web UIs and Colab notebooks for open-source voice cloning, TTS, and voice conversion models including Bark, GPT-S… | 64 | 2946 | active |
| bghira/SimpleTuner SimpleTuner is a Python fine-tuning toolkit for image, video, and audio diffusion models built on Hugging Face Diffusers. It provides a web… | 93 | 2945 | active |
| Kruk2/jasna Jasna is a GPU-accelerated tool that detects and restores mosaics in JAV videos and still images, with a native GUI, CLI, and streaming sup… | 83 | 2943 | active |
| Luce-Org/lucebox Lucebox is an open-source LLM inference engine written in C++ with custom CUDA/HIP kernels, focused on speculative decoding and speculative… | 60 | 2941 | active |
| facebookresearch/omnilingual-asr An open-source multilingual speech recognition library from Meta AI supporting over 1,600 languages, including hundreds never previously co… | 47 | 2941 | active |
| sokrypton/ColabFold ColabFold is a set of tools and Google Colab notebooks that make protein structure prediction with AlphaFold2, AlphaFold3, ESMFold, and rel… | 92 | 2940 | active |
| zjunlp/EasyEdit EasyEdit is an easy-to-use knowledge editing framework for large language models, supporting methods that modify model parameters or steer … | 74 | 2929 | active |
| InternLM/InternLM-XComposer InternLM-XComposer is a family of large vision-language models from the InternLM team, including XComposer2.5 for long-context multimodal u… | 35 | 2927 | active |
| state-spaces/s4 Official implementations of Structured State Space (S4) sequence models and related variants like S4D, HiPPO, and SaShiMi, built in PyTorch… | 32 | 2927 | active |
| keras-team/keras-tuner KerasTuner is a hyperparameter optimization framework for Keras models with a define-by-run syntax for configuring search spaces. It ships … | 46 | 2923 | active |
| decoderesearch/circuit-tracer A Python library for circuit tracing and attribution graphs in language model interpretability, based on cross-layer MLP transcoders. It co… | 77 | 2918 | active |
| eric-mitchell/direct-preference-optimization A reference implementation of Direct Preference Optimization (DPO) for training language models from human preference data, built on Huggin… | 29 | 2916 | stable |
| elixir-nx/nx Nx is an Elixir library providing multi-dimensional arrays (tensors) and numerical definitions with pluggable backends. The monorepo also i… | 98 | 2915 | active |
| AllenNLP AllenNLP is an open-source NLP research library built on PyTorch for developing state-of-the-art deep learning models on linguistic tasks. … | 10 | 11886 | maintenance |
| learnables/learn2learn learn2learn is a PyTorch library for meta-learning research, providing utilities for few-shot task creation, high-level wrappers for algori… | 45 | 2894 | active |
| RL Baselines3 Zoo RL Baselines3 Zoo is a training framework for reinforcement learning agents built on Stable Baselines3. It provides CLI scripts for trainin… | 88 | 2892 | active |
| explosion/thinc Thinc is a lightweight deep learning library from the makers of spaCy offering a type-checked, functional-programming API for composing neu… | 74 | 2891 | active |
| KellerJordan/Muon Muon is a PyTorch optimizer for the hidden layers of neural networks, based on orthogonalized momentum updates via Newton-Schulz iteration.… | 56 | 2887 | active |
| Trusted-AI/AIF360 AI Fairness 360 is an open-source Python and R library from IBM Research providing fairness metrics, explanations, and bias mitigation algo… | 58 | 2872 | active |
| openmv/openmv OpenMV is an open-source machine vision platform consisting of camera hardware firmware programmable in Python 3 (MicroPython). The firmwar… | 93 | 2871 | active |
| werner-duvaud/muzero-general A commented and documented PyTorch implementation of DeepMind's MuZero reinforcement learning algorithm, designed to be easily adapted to a… | 32 | 2870 | active |
| numz/ComfyUI-SeedVR2_VideoUpscaler The official ComfyUI integration of ByteDance's SeedVR2 model for high-quality video and image upscaling, provided as custom nodes. It can … | 49 | 2866 | active |
| physical-superintelligence-lab/Psi0 Psi-Zero (Ψ₀) is an open vision-language-action (VLA) foundation model for dexterous humanoid loco-manipulation, combining a Qwen3-VL backb… | 59 | 2859 | active |
| linto-ai/whisper-timestamped A Python library extending OpenAI's Whisper models to produce accurate word-level timestamps and confidence scores during multilingual spee… | 77 | 2855 | active |
| OpenGVLab/InternImage InternImage is a large-scale CNN-based vision foundation model that uses deformable convolutions as its core operator, released with pretra… | 25 | 2854 | stable |
| PythonOT/POT POT is an open-source Python library providing a large set of differentiable solvers for optimal transport problems, including exact and re… | 85 | 2846 | stable |
| huggingface/nanotron Nanotron is a minimalistic Python library from Hugging Face for pretraining large language models with 3D parallelism (data, tensor, and pi… | 63 | 2843 | active |
| huggingface/setfit SetFit is a Python library for efficient, prompt-free few-shot fine-tuning of Sentence Transformers for text classification. It achieves hi… | 88 | 2842 | active |
| rednote-machine-learning/RedKnot RedKnot is a long-context LLM inference acceleration library built on SGLang, using head-classified KV reuse, offline KV storage with RoPE … | 57 | 2840 | active |
| google-research/kubric Kubric is a data generation pipeline from Google Research for creating semi-realistic synthetic multi-object videos with rich annotations l… | 57 | 2839 | active |
| dscripka/openWakeWord openWakeWord is an open-source Python library for detecting wake words (or phrases) in streaming audio, with pre-trained models for common … | 46 | 2837 | active |
| WhyLogs whylogs is an open-source Python data logging library for machine learning models and data pipelines. It profiles datasets with privacy-pre… | 34 | 2835 | active |
| AdapterHub Adapters is a Python add-on library for HuggingFace Transformers that integrates 10+ parameter-efficient fine-tuning methods (bottleneck ad… | 74 | 2831 | active |
| VieNeu-TTS VieNeu-TTS is an on-device Vietnamese text-to-speech library with instant zero-shot voice cloning from short reference clips, supporting bi… | 84 | 2828 | active |
| antirez/h3.c A native C inference engine for the MiniMax H3 model on Apple Silicon, using Metal for GPU acceleration. It generates video (and audio) fro… | 53 | 2828 | active |
| datachain-ai/datachain DataChain is a Python library that turns unstructured files in S3, GCS, and Azure into typed, versioned datasets queryable at warehouse spe… | 88 | 2827 | active |
| facebook/Ax Ax is a Python platform for adaptive experimentation that uses machine learning (Bayesian and bandit optimization via BoTorch) to efficient… | 93 | 2826 | active |
| Lepton AI LeptonAI is a Python library and `lep` CLI for operating NVIDIA DGX Cloud Lepton, a platform that unifies global GPU compute for AI develop… | 96 | 2824 | active |
| OpenDCAI/DataFlex DataFlex is a data-centric training framework built on LLaMA-Factory that dynamically schedules LLM training data during optimization. It p… | 63 | 2824 | active |
| Camb-ai/MARS5-TTS MARS5 is an open-source English text-to-speech model from CAMB.AI that uses a two-stage AR-NAR pipeline to generate expressive speech with … | 15 | 2821 | active |
| huawei-noah/HEBO A research monorepo from Huawei Noah's Ark Lab containing official implementations of Bayesian optimization, reinforcement learning, and ge… | 67 | 2812 | active |
| kpu/kenlm KenLM is a C++ toolkit for estimating, filtering, and querying n-gram language models with modified Kneser-Ney smoothing. It is designed fo… | 34 | 2812 | stable |
| kha-white/manga-ocr Manga OCR is a Python library providing optical character recognition for Japanese text, focused on Japanese manga. It uses a custom end-to… | 77 | 2809 | stable |