Pytorch: open-source projects on GitHub
98 repositories on Pytorch, picked by the REDDYX AI agent. Each has a summary, key features and a breakdown. Sorted by stars.
// RELATED TOPICS
🤗 Transformers
A framework for defining machine learning models in text, vision, audio, and multimodal models, for inference and training.
★ 160,482 · PythonNODE-BASED DIFFUSION MODEL ENGINE
ComfyUI is a modular engine for content creation based on diffusion models with a node-based interface. It supports generation of images, video, 3D, audio, and text local
★ 146,848 · PythonBuilding LLM from Scratch
Step-by-step implementation of a ChatGPT-like model in PyTorch: tokenizer, architecture, training, and text generation.
★ 93,387 · Jupyter NotebookHigh-performance LLM engine
A high-performance and memory-efficient engine for inference and serving large language models.
★ 79,783 · PythonYOLOV5: OBJECT DETECTION ON PYTORCH
Computer vision model for object detection, instance segmentation, and image classification. Runs on PyTorch, supports export to CoreML and ONNX.
★ 58,278 · PythonDeepSpeed: optimization library for training
DeepSpeed is an optimization deep learning library ensuring efficient distributed training and inference. Supports ZeRO, MoE and parallelism for training models with hund
★ 43,332 · PythonRay: AI Compute Engine
Ray is a distributed engine for AI computing, combining a runtime core and a set of libraries to accelerate ML workloads.
★ 42,516 · PythonDIFFUSION MODELS LIBRARY FOR GENERATION
Library for inference and training of diffusion models generating images, audio, and 3D structures. Provides ready-to-use pipelines, noise schedulers, and pretrained mode
★ 34,917 · PythonHaystack: AI Framework for LLM
An open framework for building production-ready LLM applications with modular control over search, routing, memory, and generation.
★ 25,289 · MDXHIGH-PERFORMANCE NEURAL NETWORK INFERENCE FRAMEWORK FOR MOBILE
ncnn is a neural network inference framework with no third-party dependencies for mobile, embedded, and desktop platforms. Supports CPU and Vulkan GPU backends, includes
★ 24,068 · C++OPEN-SOURCE FINANCIAL LLMS FOR DATA ANALYSIS
FinGPT is a suite of open-source financial large language models from AI4Finance Foundation. The project offers lightweight fine-tuning (under $300) instead of full train
★ 21,591 · Jupyter NotebookDATASETS - HUGE DATABASE FOR AI
The largest platform with ready-made datasets for training AI models, offering fast and convenient tools for working with data.
★ 21,583 · PythonPEFT: Efficient AI Training
The PEFT library provides state-of-the-art parameter-efficient fine-tuning methods for models, reducing resources and training time.
★ 21,199 · PythonML-ENGINEERING OPEN BOOK
Open book on ML engineering with practical materials and examples in Python.
★ 19,717 · PythonDocsGPT: Private AI Assistant
Private AI platform for agents, assistants, and enterprise search with support for multiple models.
★ 17,882 · PythonFunASR — Speech Recognition
Industrial speech recognition tool: 170× realtime, 50+ languages, diarization, emotion detection, streaming, and OpenAI‑compatible API.
★ 16,235 · PythonROBUST RL ON PYTORCH
PyTorch version of Stable Baselines with verified implementations of reinforcement learning algorithms. 13.4K stars on GitHub.
★ 14,015 · PythonRestoration of Claude Mythos
Theoretical reconstruction of the Claude Mythos architecture, built from scratch based on available scientific literature.
★ 12,420 · PythonLUDWIG - LOW-CODE AI FRAMEWORK
Build LLMs and neural networks without code. Simplicity for custom AI models included!
★ 11,791 · PythonSoup: Train LLM Easily
Soup fine-tunes an LLM from a single YAML using a streaming layer — enabling an 8B model to run on a laptop with 4 GB VRAM.
★ 10,552 · PythonOUMI EASY LLM TUNING
OUMI lets you quickly fine‑tune (SFT/RL), evaluate, and deploy Qwen, Gemma, and any open agentic LLM/VLM without extra complexity.
★ 9,433 · PythonXINFERENCE - ANY LLM ONE LINE
Replace GPT with any open model by changing just one line of code. Run LLM, multimodal, and speech models via a single production-ready API.
★ 9,318 · PythonLMCache: LLM Acceleration
The fastest KV cache layer for large language models, boosting performance and reducing latency.
★ 8,292 · PythonBITSANDBYTES: LLM COMPRESSION
Library for k-bit quantization of large language models, enabling LLM execution on regular hardware via PyTorch.
★ 8,246 · PythonSenseVoice: Voice Understanding Model
Multilingual model for analyzing and understanding speech in different languages with high accuracy.
★ 8,171 · PythonLLM FROM SCRATCH IN PYTORCH
The reasoning-from-scratch project shows how to build a Reasoning LLM from scratch in PyTorch, step by step. Implemented in Jupyter notebooks: architecture, training with
★ 5,831 · Jupyter NotebookDOCKER WRAPPER KOKORO-82M TTS WITH OPENAI API
Docker image with FastAPI wrapper over Kokoro-82M speech synthesis model. Provides OpenAI-compatible API, websockets for voice agents, and web UI for audio generation.
★ 5,706 · PythonZENML: AI PLATFORM FROM PIPELINE
ZenML is a Python framework for creating AI pipelines and agents, unifying data, models, and orchestration in a single platform.
★ 5,636 · PythonMMF: Modular Multimodal Framework
Modular framework from FAIR for vision and language research, combining the flexibility of Python and the power of multimodal models.
★ 5,633 · PythonEfficient Multimodal Inference
Framework for efficient inference of multimodal models. Unifies text, images, and audio in a single generation system.
★ 4,863 · PythonVLMEvalKit - TOP TOOL FOR MODEL EVALUATION
Open tool for evaluating multimodal models. Supports 220+ LMM and 80+ benchmarks.
★ 4,492 · PythonSWANLAB: TRACKING AND VISUALIZATION
SwanLab is an open-source tool for tracking and visualizing AI model training, compatible with PyTorch, Transformers, and other frameworks.
★ 3,981 · PythonAI‑ENGINEER HQ: SCIENCE CORE
The project contains Jupyter notebooks with methods for training models, RAG pipelines on LangChain and data structure algorithms for AI agents. It allows quickly creatin
★ 3,780 · Jupyter NotebookCHITU: ACCELERATING LLMS IN PYTHON
High-performance framework for fast inference of large language models in Python. Provides flexibility, efficiency, and ease of deployment.
★ 3,123 · Pythonrl-baselines3-zoo - RL Training with Hyperparameters
Framework for training Stable Baselines3 agents with automatic hyperparameter tuning and ready pre-trained models.
★ 2,912 · PythonAUTODISTILL: TRAINING WITHOUT LABELING
Automatically turns raw images into ready-to-use inference models, using foundation models to generate pseudo‑labels and train supervised models. No manual labeling requi
★ 2,798 · PythonTrainer: Distributed Training
The trainer project enables distributed training of AI models and fine‑tuning of LLMs in Kubernetes, written in Go.
★ 2,284 · GoHands-on Modern RL
An open course on reinforcement learning: from basic algorithms to LLM alignment, RLVR, and agentic systems.
★ 1,850 · PythonSGLANG-OMNI: OMNI-SERVICE FOR AI
SGLang-Omni provides high-performance execution of TTS, ASR, and speech omni-models on Python.
★ 1,791 · PythonKATIB DEPRECATED: MIGRATE TO TRAINER
Project Katib is deprecated. Migrate to OptimizationJob and Kubeflow Trainer for hyperparameter tuning.
★ 1,708 · PythonTOOLKIT FOR CLINICAL AI
PyHealth is a library for healthcare AI supporting EHR (MIMIC, eICU, OMOP), clinical texts, signals, and imaging. 30+ models and reproducible pipelines for research.
★ 1,699 · PythonPARALLAX: BUILD AI CLUSTER
Parallax lets you quickly deploy a distributed inference service on any infrastructure.
★ 1,301 · PythonQUICKLY TRAIN SPECULATIVE DECODER
SpecForge simplifies training speculative decoding models and exporting them to SGLang for high-performance serving.
★ 1,285 · PythonKVPRESS: LLM CACHE COMPRESSION
Library for easy compression of LLM KV‑cache, accelerating generation and reducing GPU memory consumption.
★ 1,254 · PythonMULTIMODAL AI FOR CONTENT
UForm — compact MULTIMODAL AI for understanding and generating content. Processes multilingual texts, images and video, working 5x faster than OpenAI CLIP and LLaVA.
★ 1,253 · PythonRECOMMENDATIONS ON STEROIDS
A framework for recommendation models on PyTorch. Easy to use and easily extensible.
★ 1,242 · Jupyter NotebookNNCFF - ACCELERATING MODEL INFERENCE
NNCFF is a neural network compression framework for accelerated inference in OpenVINO™. It reduces model size while preserving accuracy.
★ 1,221 · PythonAI-HANDS-ON: LEARN AI FROM SCRATCH
A collection of notebooks and files for learning AI from scratch. 1137 stars on GitHub.
★ 1,137 · Jupyter NotebookAI Agent 24/7
An autonomous AI agent runs your deep experiments 24/7, saving resources and memory. The Leader‑Worker architecture ensures stable operation at zero cost.
★ 1,035 · PythonTUTEL ACCELERATES MOE IN C
Optimized Mixture-of-Experts library for accelerated LLM inference with FP8, NVFP4, and MXFP4 support.
★ 1,028 · CInferenceX – Inference Benchmark
Open-source continuous inference benchmark for Qwen3.5, DeepSeek, GPTOSS and new accelerators.
★ 1,016 · PythonKeras Hub: Models for Keras
Platform with ready-to-use pretrained models for Keras 3, simplifying AI integration into projects.
★ 999 · PythonHow to Train Your LLM
Build a modern LLM from scratch — every line of code is commented and explained simply, as if for a five-year-old.
★ 984 · Jupyter NotebookAgileRL - Evolutionary Acceleration
AgileRL speeds up RL training 10× via evolutionary hyperparameter optimization and ready-to-use SOTA algorithms.
★ 967 · PythonMOSEC - MAXIMUM ML PRODUCTIVITY
Fast framework for serving ML models with dynamic batching and CPU/GPU pipelines.
★ 904 · PythonHALO: FRAMEWORK FOR LLM TRAINING
Halo is an open-source framework from White Circle for training large language and multimodal models. It supports agent-based training, async-RL, distributed training, an
★ 862 · PythonML from First Principles
Educational resource teaching machine learning concepts from first principles, covering inference, CUDA, deep learning, and reinforcement learning with Python implementat
★ 842 · PythonSTABLE-BASELINES3-CONTRIB: NEW RL ALGORITHMS
The contrib package for Stable-Baselines3 contains experimental reinforcement learning algorithms. It includes GSDE implementations, Gym environment support, and PyTorch
★ 746 · PythonLOCAL AI AUDIO GENERATION
audio-studio-mcp runs Stable Audio 3 on Apple Silicon via MLX and on NVIDIA GPU via PyTorch. Generates background music and sound effects for Claude Code and any MCP clie
★ 669 · PythonATORCH: Pythonic PyTorch
The attorch library offers a subset of PyTorch neural network modules implemented in Python using OpenAI's Triton for accelerated computation.
★ 607 · PythonNew projects on this topic in Telegram
Telegram →