Ecosystem & Stack: pytorch

Unsloth Studio is a web UI that enables efficient local training and inference of open-source large language models and other AI models with significant VRAM and speed optimizations.

A scalable and lightweight infrastructure for fine-tuning, inference, and deployment of over 1000 large language models (LLMs) and multimodal large language models (MLLMs) using advanced techniques.

SGLang is a high-performance serving framework for large language models and multimodal models, optimizing inference throughput and latency.

A state-of-the-art library for Parameter-Efficient Fine-Tuning (PEFT) of large pretrained models, drastically reducing computational and storage costs.

vLLM is a high-throughput and memory-efficient open-source library designed for fast and easy serving of large language models.

Streamlines complex deep learning engineering, enabling scalable AI model training and finetuning across diverse hardware with minimal code changes.

A next-generation training engine optimized for ultra-large Mixture-of-Experts (MoE) models, offering superior efficiency and scalability.

An educational project providing step-by-step code to build a ChatGPT-like Large Language Model (LLM) from scratch using PyTorch.

Heretic is an AI model utility that automatically removes censorship and safety alignment from transformer-based language models without requiring expensive post-training.

A comprehensive, job-oriented guide for AI Agent development, covering core technologies, practical projects, and interview preparation for LLM-related roles.

A KVCache-centric disaggregated architecture for high-performance LLM serving, powering leading AI services.

SwanLab is an open-source, modern-design platform for tracking, visualizing, and analyzing AI/ML training experiments, supporting cloud and self-hosted deployments.

A Kubernetes-native platform for scalable distributed AI model training and LLM fine-tuning across various frameworks.

An open collection of methodologies, tools, and step-by-step instructions for successful training, fine-tuning, and inference of large language and multi-modal models.

Deep Lake is an AI data runtime and database optimized for deep learning, offering serverless multimodal data storage, scalable retrieval, and training capabilities.

An open-source, cloud-native, all-in-one MLOps platform designed for the full lifecycle management of machine learning, deep learning, and large language model development and deployment.

A PyTorch library enabling accessible large language models through k-bit quantization, significantly reducing memory consumption for both inference and training.

A low-code, declarative framework for building and deploying custom large language models (LLMs) and other deep neural networks with ease and efficiency.

OpenVINO is an open-source toolkit designed to optimize and deploy deep learning models for efficient AI inference across a wide range of hardware platforms.

A 12-week, 24-lesson curriculum from Microsoft to learn Artificial Intelligence for beginners, including practical lessons, quizzes, and labs.

A lightweight library providing one-line dataloaders and efficient pre-processing tools for a vast hub of AI datasets, supporting various ML frameworks.

A Python library for representing, transmitting, storing, and retrieving multimodal data, designed for AI applications.

xTuring simplifies the process of fine-tuning and deploying open-source Large Language Models (LLMs) on private data, ensuring privacy, efficiency, and scalability.

A PyTorch-based framework to adapt Meta AI's Segment Anything Model (SAM) for improved performance on challenging downstream computer vision tasks using adapters and prompts.

A toolkit for quantizing (compressing) Large Language Models (LLMs) with hardware acceleration across various GPUs and CPUs, integrating with popular inference frameworks.

A deep learning toolkit for training custom multimodal large language models focused on speech, language, audio, and music processing.

A comprehensive code repository accompanying a hands-on course for mastering Huggingface Transformers, covering fundamental concepts to advanced fine-tuning and deployment techniques.

A beginner-friendly UI for fine-tuning large language models (LLMs) using the LoRA method on commodity NVIDIA GPUs.

A cost-effective solution for finetuning ChatGLM-6B with LoRA, enabling personalized large language models.

A unified library extending HuggingFace Transformers for parameter-efficient and modular transfer learning in NLP.

Optimizes large language model inference to run 70B models on a single 4GB GPU without quantization, enabling efficient deployment on resource-constrained hardware.

A comprehensive collection of PyTorch implementations for over 60 deep learning papers, featuring side-by-side annotated notes for enhanced understanding.

Provides Google Colab notebooks for easily deploying and running Stable Diffusion WebUI, enabling AI-powered image generation and training without local hardware.

A Python library implementing LoRA (Low-Rank Adaptation) to efficiently fine-tune large language models by significantly reducing trainable parameters and storage requirements.

BELLE is an open-source project dedicated to fostering the development of Chinese conversational large language models, aiming to make LLMs accessible to everyone.

YaYi is an open-source Chinese Large Language Model, built on LLaMA 2 & BLOOM, designed for secure, reliable, and domain-specific applications through extensive instruction tuning.

A modular framework for aligning any-modality large models with human intentions and values using various fine-tuning and reinforcement learning methods.

A human preference reward model for evaluating and improving text-to-image generation models.

A comprehensive collection of recipes and code for training various reward models crucial for Reinforcement Learning from Human Feedback (RLHF) in large language models.

An open-source framework providing code, models, and insights for stable Reinforcement Learning from Human Feedback (RLHF) training in Large Language Models, focusing on the PPO algorithm and reward modeling.

A modular PyTorch library for state-of-the-art diffusion models, enabling easy inference and training for image, video, and audio generation.

An open-source, AI-driven tool for advanced image inpainting, outpainting, object removal, and replacement using state-of-the-art models.

Integrates Stable Diffusion directly into Blender for seamless AI-powered texture generation, concept art creation, and image manipulation within 3D workflows.

A PyTorch implementation for generating 3D models from text or images, leveraging NeRF and diffusion models like Stable Diffusion.

A fast, minimal PyTorch port of DALL·E Mini for efficient text-to-image generation.

A PyTorch implementation of Google's Imagen, a state-of-the-art text-to-image neural network, enabling advanced generative AI capabilities.

A PyTorch implementation of OpenAI's DALL-E 2, a state-of-the-art neural network for text-to-image synthesis.

An open-source PyTorch implementation and replication of OpenAI's DALL-E, a text-to-image transformer, including CLIP for generation ranking.

This project implements Google's Dreambooth technique on Stable Diffusion, enabling users to fine-tune a text-to-image model with a few custom examples for personalized image generation.

A scalable generative AI framework for researchers and developers focused on Large Language Models, Multimodal, and Speech AI (ASR, TTS).

A collection of pre-trained, end-to-end text-to-speech models designed for simplicity, speed, and natural-sounding speech across multiple languages.

An open-source text-to-speech model that combines the VITS2 backbone with multilingual BERT for high-quality, multi-language speech synthesis.

Generate high-quality audiobooks and voiceovers from various text formats with synchronized captions.

A real-time voice cloning toolkit that allows users to replicate a voice in 5 seconds and generate arbitrary speech.

A Python-based tool to convert e-books (EPUB) into high-quality M4B audiobooks using advanced text-to-speech models.

A powerful web-based tool for few-shot voice cloning and text-to-speech, enabling high-quality voice generation from minimal audio data.

A Dockerized FastAPI wrapper for the Kokoro-82M text-to-speech model, offering multi-language support, CPU/GPU inference, and an OpenAI-compatible API.

A declarative, transactional Python library for building multimodal AI applications with incremental data storage, transformation, indexing, and orchestration.

A unified, reproducible, and efficient multimodal evaluation toolkit for large language models across text, image, video, and audio tasks.

An open-source AI model family for high-fidelity, expressive speech and sound generation across diverse real-world applications.

A PyTorch implementation of BitNet, enabling highly efficient 1-bit transformers for large language models.

A modular and scalable PyTorch-based framework for state-of-the-art vision and language multimodal research from Facebook AI Research.

A Python library for extracting clean markdown, multimodal media, and structured data from complex documents using vision-language models.

A high-performance, comprehensive face recognition library built on PaddlePaddle and PyTorch.

An open-source PyTorch-based framework for NLP pre-training and fine-tuning, offering modularity, reproducibility, and a comprehensive model zoo for various downstream tasks.

A toolkit for finetuning ChatGLM series models (ChatGLM-6B, ChatGLM2-6B, ChatGLM3-6B) using various methods like Freeze, Lora, P-tuning, and full parameter training for downstream NLP tasks.

A unified platform simplifying instruction-tuning for Large Language Models by integrating diverse data, LLMs, and parameter-efficient methods.

An open-source AI model for instant, accurate, and flexible voice cloning, supporting cross-lingual synthesis and granular style control.

DiffSinger is an official PyTorch implementation of a singing voice synthesis (SVS) and text-to-speech (TTS) system, leveraging a shallow diffusion mechanism for high-quality audio generation.

EmotiVoice is an open-source, multi-voice, and prompt-controlled text-to-speech engine supporting English and Chinese with emotional synthesis capabilities.

StyleTTS 2 is a cutting-edge text-to-speech model achieving human-level speech synthesis through style diffusion and adversarial training with large speech language models.

MetaVoice-1B is an open-source 1.2B parameter foundational model for highly expressive, human-like text-to-speech synthesis with advanced voice cloning capabilities.

An open-source implementation of Microsoft's VALL-E X, enabling zero-shot multilingual text-to-speech synthesis and voice cloning with emotion control.

A deep learning library for advanced, high-quality, and efficient Text-to-Speech (TTS) synthesis, supporting multiple languages and models.

LyCORIS is a library implementing various parameter-efficient fine-tuning (PEFT) algorithms for Stable Diffusion, extending beyond conventional LoRA methods to enhance model adaptation.

Create dynamic and visually captivating videos by smoothly morphing between different text prompts using Stable Diffusion.

Official SeedVR2 Video Upscaler for ComfyUI, enabling high-quality video and image upscaling, also runnable as a standalone CLI.

An open-source, theoretical reconstruction of the Claude Mythos LLM architecture, featuring a Recurrent-Depth Transformer and sparse Mixture of Experts for advanced reasoning.

A user-friendly template integrating PyTorch Lightning and Hydra to streamline deep learning experimentation and development.

A lightweight and optimized Python library for fast offline large language model inference, offering comparable or better performance than vLLM with a more readable codebase.

AdalFlow is a PyTorch-like open-source library designed to build and automatically optimize large language model (LLM) applications, from chatbots and RAG systems to complex AI agents.

A high-performance, no-abstraction toolkit providing recipes for pretraining, finetuning, and deploying over 20 large language models at scale.

MMPreTrain is an OpenMMLab project providing a comprehensive, open-source PyTorch-based toolbox for pre-training and benchmarking various computer vision and multi-modal models.

A PyTorch library providing advanced foundation architectures to efficiently and effectively scale Transformers for large language models and general-purpose AI.

Create stunning Disco Diffusion artworks with a single line of Python code, offering a professional API and robust integration capabilities.

A library for real-time music and audio generation leveraging stable diffusion, offering CLI, interactive app, and API capabilities.

MOSS-TTS-Nano is an open-source, multilingual, tiny speech generation model optimized for real-time CPU inference and lightweight integration.