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 |
|---|---|---|---|
| QwenLM/Qwen3-Embedding Qwen3-Embedding is a series of text embedding and reranking models (0.6B, 4B, 8B) built on Qwen3 foundation models, with a Python repositor… | 35 | 2040 | active |
| pykeen/pykeen PyKEEN is a Python library for training and evaluating multimodal knowledge graph embedding models built on PyTorch. It provides a high-lev… | 69 | 2039 | active |
| PRIS-CV/DemoFusion DemoFusion is a CVPR 2024 framework that extends open-source latent diffusion models like SDXL to generate high-resolution images without a… | 45 | 2038 | stable |
| tensorflow/privacy TensorFlow Privacy is a Python library providing TensorFlow optimizers for training machine learning models with differential privacy. It i… | 64 | 2036 | active |
| SakanaAI/continuous-thought-machines The Continuous Thought Machine (CTM) is a neural network architecture from Sakana AI that uses neuron-level temporal dynamics and neural sy… | 43 | 2036 | active |
| tensorflow/recommenders TensorFlow Recommenders is a Python library for building recommender system models on top of TensorFlow and Keras. It covers the full workf… | 84 | 2034 | active |
| Anime4KCPP Anime4KCPP is a high-performance anime image and video upscaler built on CNN-based algorithms, written in C++. It ships as a library plus V… | 72 | 2033 | active |
| languine-ai/languine Languine is an AI-powered localization CLI and self-hostable platform that automatically translates application i18n files into multiple la… | 56 | 2033 | active |
| deep-floyd/IF DeepFloyd IF is an open-source text-to-image model library implementing a cascaded pixel diffusion architecture with a frozen T5 text encod… | 22 | 7793 | maintenance |
| bytetriper/RAE Official PyTorch implementation of 'Diffusion Transformers with Representation Autoencoders' (RAE), a two-stage image generation pipeline u… | 45 | 2032 | active |
| tdrussell/diffusion-pipe A Python training script for fine-tuning diffusion models (image and video generation) using DeepSpeed pipeline parallelism across multiple… | 67 | 2028 | active |
| ZhuJHua/moodiary Moodiary is a fully open-source, cross-platform diary/journaling app built with Flutter and Rust. It supports markdown, plain text, and ric… | 86 | 2027 | active |
| D-Ogi/WatermarkRemover-AI An AI-powered desktop application that detects and removes watermarks from images and videos using Microsoft's Florence-2 for detection and… | 77 | 2027 | active |
| Tencent-Hunyuan/HunyuanOCR HunyuanOCR-1.5 is a lightweight end-to-end OCR vision-language model from Tencent, with a unified inference environment, llama.cpp PC-side … | 60 | 2025 | active |
| NUS-HPC-AI-Lab/VideoSys VideoSys is an open-source Python library providing easy and efficient infrastructure for video generation, supporting training, inference,… | 30 | 2025 | active |
| instantX-research/InstantStyle InstantStyle is a framework for style-preserving text-to-image generation that disentangles style and content from reference images using f… | 27 | 2019 | active |
| XavierXiao/Dreambooth-Stable-Diffusion An implementation of Google's Dreambooth fine-tuning method applied to Stable Diffusion, enabling personalization of a text-to-image diffus… | 32 | 7730 | maintenance |
| tatsu-lab/alpaca_eval AlpacaEval is an LLM-based automatic evaluation framework for instruction-following language models, producing win rates against a GPT-4 ba… | 33 | 2018 | active |
| NVlabs/SPADE Official PyTorch implementation of SPADE (GauGAN), a CVPR 2019 method for synthesizing photorealistic images from semantic segmentation map… | 32 | 7717 | maintenance |
| uncertainty-toolbox/uncertainty-toolbox A Python library for predictive uncertainty quantification, providing metrics, visualizations, and recalibration procedures for regression … | 23 | 2015 | active |
| UniFace UniFace is a unified Python library for face analysis that bundles detection, recognition, landmark localization, face parsing, gaze estima… | 85 | 2012 | active |
| meta-recsys/generative-recommenders Meta's research library implementing HSTU and M-FALCON from the ICML'24 paper 'Actions Speak Louder than Words: Trillion-Parameter Sequenti… | 70 | 2012 | active |
| philipperemy/keras-tcn A Keras/TensorFlow implementation of Temporal Convolutional Networks (TCN) with dilated causal convolutions, usable as a drop-in layer alte… | 58 | 2011 | active |
| cambrian-mllm/cambrian Cambrian-1 is a fully open family of vision-centric multimodal large language models (MLLMs) from NYU's VISIONx group, with training and ev… | 43 | 2011 | active |
| lyhue1991/torchkeras torchkeras is a lightweight PyTorch model training template library that brings Keras-style compile/fit/evaluate APIs to PyTorch. Its core … | 52 | 2010 | active |
| AntixK/PyTorch-VAE A collection of Variational Autoencoder (VAE) model implementations in PyTorch, including Beta-VAE, VQ-VAE, IWAE, WAE, and others, with a f… | 34 | 7678 | maintenance |
| FireRedASR2S FireRedASR is a family of open-source industrial-grade automatic speech recognition models supporting Mandarin, Chinese dialects, and Engli… | 49 | 2005 | active |
| mil-tokyo/webdnn WebDNN is a framework for running deep neural network inference directly in the web browser, accepting ONNX models without Python preproces… | 68 | 2003 | active |
| chaidiscovery/chai-lab Chai-1 is a state-of-the-art multi-modal foundation model for biomolecular structure prediction, handling proteins, small molecules, DNA, R… | 62 | 2001 | active |
| zai-org/GLM-130B GLM-130B is an open bilingual (English and Chinese) 130-billion-parameter dense language model pre-trained with the General Language Model … | 32 | 7650 | maintenance |
| clementchadebec/benchmark_VAE Pythae is a PyTorch library that unifies implementations of many Variational Autoencoder (VAE) variants under a common interface, enabling … | 23 | 2000 | active |
| elyra-ai/elyra Elyra is a set of AI-centric extensions for JupyterLab, including a visual pipeline editor for building and executing notebook-based pipeli… | 67 | 1998 | active |
| lucidrains/titans-pytorch An unofficial PyTorch implementation of the Titans architecture, a neural long-term memory module for transformers that learns to memorize … | 64 | 1996 | active |
| allenai/scispacy scispaCy is a Python library providing full spaCy pipelines, models, and custom pipes for processing scientific, biomedical, and clinical t… | 47 | 1994 | active |
| logpai/logparser Logparser is a Python machine learning toolkit and benchmark suite for automated log parsing. It extracts event templates from unstructured… | 30 | 1994 | active |
| hkchengrex/XMem XMem is a PyTorch model for semi-supervised video object segmentation that tracks objects through long videos using an Atkinson-Shiffrin-in… | 23 | 1994 | stable |
| facebookresearch/dino PyTorch implementation of DINO, a self-supervised learning method for training Vision Transformers, with pretrained model weights. It is th… | 10 | 7609 | maintenance |
| kwuking/TimeMixer Official PyTorch implementation of TimeMixer, an ICLR 2024 model for time series forecasting using decomposable multiscale mixing. It has s… | 42 | 1991 | active |
| featureform/featureform Featureform is a virtual feature store that sits atop your existing data infrastructure and orchestrates it to define, manage, and serve ML… | 32 | 1991 | active |
| jianchang512/vocal-separate A minimal local web-based tool for separating vocals from background music in audio or video files, using Spleeter 2stems/4stems/5stems mod… | 10 | 1990 | active |
| haykgrigo3/TimeCapsuleLLM TimeCapsuleLLM is a research project training language models from scratch (and fine-tuning small base models) exclusively on text from spe… | 70 | 1989 | active |
| Hyperopt Hyperopt is a Python library for distributed asynchronous hyperparameter optimization over search spaces with real-valued, discrete, and co… | 84 | 7595 | maintenance |
| GPT2-Chinese A Python library providing GPT-2 training and text generation code tailored for Chinese, built on HuggingFace Transformers with BERT or BPE… | 32 | 7594 | maintenance |
| davda54/sam An unofficial PyTorch implementation of Sharpness-Aware Minimization (SAM) and its adaptive variant ASAM, provided as an optimizer wrapper … | 32 | 1986 | stable |
| xingyizhou/CenterNet CenterNet is a PyTorch implementation of the 'Objects as Points' detector, which models objects as single center points detected via keypoi… | 32 | 7574 | maintenance |
| android/androidify An open-source Android sample app from Google that lets users create custom Android bot avatars using AI image generation via the Gemini AP… | 66 | 1983 | active |
| Netflix/void-model VOID (Video Object and Interaction Deletion) is a research model from Netflix that removes objects from videos along with the physical inte… | 51 | 1983 | active |
| patrikhuber/eos A lightweight, header-only 3D Morphable Face Model (3DMM) fitting library written in modern C++11/14, with Python bindings. It provides mod… | 28 | 1981 | active |
| cloneofsimo/lora A Python library for applying Low-Rank Adaptation (LoRA) to quickly fine-tune text-to-image diffusion models like Stable Diffusion. It prod… | 23 | 7557 | maintenance |
| haoheliu/versatile_audio_super_resolution AudioSR is a Python library and CLI tool that performs versatile audio super-resolution, upsampling any audio (music, speech, sound effects… | 42 | 1979 | active |
| WassimTenachi/PhySO PhySO is a Python library for physical symbolic optimization that uses deep reinforcement learning to discover analytical physical laws fro… | 49 | 1978 | active |
| showlab/Show-o Show-o is a research repository implementing a unified transformer model that combines autoregressive and discrete diffusion modeling for m… | 48 | 1978 | active |
| Linzaer/Ultra-Light-Fast-Generic-Face-Detector-1MB An ultra-lightweight face detection model (~1MB FP32, ~300KB quantized) designed for edge computing devices, with slim and RFB variants tra… | 32 | 7548 | maintenance |
| adobe-research/custom-diffusion Custom Diffusion is a research codebase for efficiently fine-tuning text-to-image diffusion models like Stable Diffusion on a few example i… | 66 | 1977 | stable |
| 2U1/Qwen-VL-Series-Finetune An open-source Python repository providing training scripts for fine-tuning Alibaba's Qwen-VL series of vision-language models (Qwen2-VL, Q… | 65 | 1977 | active |
| OpenMotionLab/MotionGPT MotionGPT is a unified motion-language model that treats 3D human motion as a foreign language by converting motion into discrete motion to… | 30 | 1974 | active |
| openlm-research/open_llama OpenLLaMA is a permissively licensed (Apache-2.0) open reproduction of Meta AI's LLaMA, releasing 3B, 7B, and 13B pretrained model weights … | 30 | 7527 | maintenance |
| jd-opensource/JoyAI-VL-Interaction JoyAI-VL-Interaction is an open 8B-scale vision-language interaction model with a complete deployable real-time streaming system, including… | 59 | 1970 | active |
| JIA-Lab-research/DreamOmni2 DreamOmni2 is the official PyTorch implementation of a CVPR 2026 Highlight model for multimodal instruction-based image editing and generat… | 48 | 1968 | active |
| StarCoder StarCoder is a 15B-parameter code language model trained on 80+ programming languages, and this repository hosts the fine-tuning and infere… | 30 | 7501 | maintenance |
| meta-pytorch/opacus Opacus is a PyTorch library for training neural networks with differential privacy via DP-SGD, requiring minimal code changes through its P… | 78 | 1967 | active |
| cra-ros-pkg/robot_localization robot_localization is a ROS package of nonlinear state estimation nodes (EKF, UKF, and navsat_transform) for fusing IMU, odometry, and othe… | 61 | 1966 | stable |
| siliconflow/onediff OneDiff is an out-of-the-box acceleration library for diffusion models, providing PyTorch compilation tools and optimized GPU kernels. It i… | 44 | 1965 | active |
| NVlabs/RADIO Official PyTorch implementation of AM-RADIO and its successors (RADIOv2.5, C-RADIOv4), agglomerative vision foundation models distilled fro… | 61 | 1961 | active |
| open-mmlab/mmagic MMagic is OpenMMLab's toolbox for generative and multimodal AI image/video creation, built on PyTorch. It provides a large model zoo coveri… | 23 | 7469 | maintenance |
| SizheAn/PanoHead PanoHead is the official PyTorch implementation of a CVPR 2023 paper presenting a 3D-aware GAN that synthesizes geometry-aware, view-consis… | 30 | 1960 | active |
| autodiff/autodiff autodiff is a C++17 library for automatic differentiation that computes derivatives of functions efficiently using forward mode (dual numbe… | 23 | 1960 | active |
| CannyLab/tsne-cuda A CUDA-accelerated implementation of the FIt-SNE t-SNE algorithm with Python bindings, offering up to 1200x speedup over scikit-learn. It e… | 89 | 1958 | stable |
| Yuliang-Liu/Monkey Monkey is a large multi-modal model (LMM) research project from CVPR 2024 that improves image understanding via higher input resolution and… | 71 | 1956 | active |
| jd-opensource/JoyAI-Video-Edit JoyAI-Video-Edit is a real-time, instruction-guided video editing system that applies natural-language edits to live or uploaded video stre… | 58 | 1956 | active |
| OminiControl OminiControl is a universal control framework for Diffusion Transformer models like FLUX, supporting subject-driven and spatial control (ed… | 66 | 1955 | active |
| alibaba/EasyCV EasyCV is an all-in-one PyTorch-based computer vision toolkit from Alibaba covering self-supervised learning, vision transformers, and majo… | 28 | 1955 | active |
| LTH14/mar Official PyTorch implementation of MAR (Masked Autoregressive) image generation with DiffLoss, from the NeurIPS 2024 paper 'Autoregressive … | 51 | 1954 | stable |
| eriklindernoren/PyTorch-YOLOv3 A minimal PyTorch implementation of YOLOv3 supporting training, inference, and evaluation, with compatibility for YOLOv4 and YOLOv7 weights… | 32 | 7439 | maintenance |
| Fafa-DL/Awesome-Backbones A PyTorch-based framework that integrates many deep learning backbone models (CNNs and vision transformers like ResNet, EfficientNet, Swin … | 32 | 1953 | active |
| openai/guided-diffusion OpenAI's codebase for guided diffusion models from the paper 'Diffusion Models Beat GANs on Image Synthesis', including classifier conditio… | 32 | 7427 | maintenance |
| kadirnar/whisper-plus A Python library wrapping OpenAI Whisper-family models (including distil-whisper and MLX variants) for fast speech-to-text transcription wi… | 62 | 1950 | active |
| deepseek-ai/TileKernels TileKernels is a Python library of optimized GPU kernels for LLM operations, written in the TileLang DSL. It provides kernels for MoE routi… | 58 | 1949 | active |
| Vchitect/Latte Official PyTorch implementation of Latte, a latent diffusion transformer for video generation. It includes model definitions, pre-trained c… | 67 | 1948 | active |
| microsoft/Magma Magma is Microsoft Research's foundation model for multimodal AI agents, released as an 8B vision-language model that understands images an… | 50 | 1948 | active |
| kyegomez/BitNet A PyTorch implementation of the BitNet architecture from the paper 'BitNet: Scaling 1-bit Transformers for Large Language Models', providin… | 73 | 1947 | active |
| starik222/BooruDatasetTagManager A desktop tag editor for managing booru-style tagged image and video datasets used to train Stable Diffusion models such as LoRAs, embeddin… | 70 | 1947 | active |
| Graph Convolutional Networks A TensorFlow implementation of Graph Convolutional Networks (GCN) for semi-supervised node classification on graphs, accompanying the ICLR … | 32 | 7406 | maintenance |
| MLJ MLJ.jl is a machine learning framework for Julia providing a common interface to over 200 models, with meta-algorithms for model selection,… | 91 | 1946 | stable |
| BytedTsinghua-SIA/DAPO DAPO is an open-source reinforcement learning system for large-scale LLM training, released by ByteDance Seed and Tsinghua AIR. It implemen… | 26 | 1946 | active |
| tensorlayer/TensorLayer TensorLayer is a TensorFlow-based deep learning and reinforcement learning library offering customizable neural layers for researchers and … | 23 | 7383 | maintenance |
| c2g-dev/city2graph City2Graph is a Python library that converts geospatial data such as buildings, streets, GTFS transit feeds, and origin-destination matrice… | 84 | 1940 | active |
| tum-pbs/PhiFlow PhiFlow is an open-source Python simulation toolkit for solving partial differential equations with support for optimization and machine le… | 63 | 1939 | active |
| PixArt-alpha/PixArt-sigma PixArt-Σ is a PyTorch implementation of a diffusion transformer model for high-resolution (up to 4K) text-to-image generation, trained with… | 26 | 1939 | active |
| Audio-AGI/AudioSep AudioSep is the official implementation of the 'Separate Anything You Describe' foundation model for open-domain, language-queried audio so… | 29 | 1938 | active |
| fan-ziqi/robot_lab robot_lab is a reinforcement learning extension library for robots built on top of NVIDIA IsaacLab, providing preconfigured locomotion envi… | 77 | 1937 | active |
| llvm/torch-mlir Torch-MLIR is a compiler project providing first-class translation of PyTorch programs into the MLIR compiler ecosystem. It lets hardware v… | 99 | 1935 | active |
| wuyoscar/Internal-Safety-Collapse ISC-Bench/TVD is a research framework for studying 'Internal Safety Collapse' in frontier LLMs, where agents placed in adversarial codespac… | 60 | 1935 | active |
| PAIR-code/facets Facets is a pair of web-component visualizations (Overview and Dive) for understanding and analyzing machine learning datasets, embeddable … | 10 | 7333 | maintenance |
| NVlabs/nvdiffrast Nvdiffrast is a PyTorch library from NVIDIA providing high-performance, GPU-accelerated primitive operations for rasterization-based differ… | 49 | 1922 | stable |
| GPflow/GPflow GPflow is a Python library for building Gaussian process models on top of TensorFlow 2 and TensorFlow Probability. It implements modern Gau… | 95 | 1921 | active |
| OpenTalker/video-retalking VideoReTalking is a Python research system from SIGGRAPH Asia 2022 that edits real-world talking-head videos to match a given audio track, … | 23 | 7292 | maintenance |
| IPADS-SAI/MobiAgent MobiAgent is a systematic framework for building customizable GUI agents that operate mobile phones, comprising the MobiMind agent model fa… | 54 | 1920 | active |
| pymatting/pymatting PyMatting is a Python library for alpha matting that estimates an alpha matte from an input image and a hand-drawn trimap to extract foregr… | 84 | 1918 | active |
| Robbyant/lingbot-va LingBot-VA is an autoregressive diffusion framework that unifies video world modeling and robot policy learning in a single interleaved vid… | 53 | 1918 | active |