Genesis: open source generative physics engine for real physics-based 4D dynamic world simulation
综合介绍 Genesis 是一个为通用机器人和具身 AI 学习设计的生成性物理世界。它提供了一个统一的仿真平台，支持多种材料和物理现象的模拟。Genesis 旨在通过结合生成性 AI 和物理仿真，解锁...
Sorting
Kolors: text-to-image model for generating high-quality images, support for generating Chinese posters
综合介绍 Kolors是由快手团队开发的大规模文本到图像生成模型，基于潜在扩散技术。该模型训练于数十亿对文本-图像数据对，能够生成高质量、复杂语义准确的图像，并支持中英文输入。Kolors在视觉质量...
ColorFlow: Comic book coloring, automatic coloring of black and white images to improve image color consistency and quality
综合介绍 ColorFlow是由腾讯ARC团队开发的图像序列自动着色工具，旨在解决黑白图像序列的自动着色问题。该工具利用检索增强的着色管道，通过参考图像池准确生成各种元素的颜色，包括角色的头发颜色和服...
BrushEdit: All-in-one tool for image restoration and editing released by Tencent ARC
Comprehensive Introduction BrushEdit is an all-in-one image repair and editing tool developed by Tencent ARC Lab. Based on the latest AI technology, the tool can automatically recognize and repair defects in images, while supporting users to make interactive edits.BrushEdit combines a variety of...
Outlines: Generate structured text output via regular expressions, JSON or Pydantic models
综合介绍 Outlines 是一个由 dottxt-ai 开发的开源库，旨在通过结构化文本生成来提升大语言模型（LLM）的应用能力。该库支持多种模型集成，包括 OpenAI、transformers...
RapBank: a model for directly generating rap (Rap) vocals from lyrics and backing tracks (currently open dataset)
General Introduction RapBank is a dataset and toolset designed for rap lyrics generation. The project was created by NZqian to provide researchers and developers with a high-quality rap lyrics data by collecting and processing rap songs from YouTube...
R2R: An Advanced AI Retrieval (RAG) System for Multimodal Content Parsing and Combining Knowledge Graph with Hybrid Search
Comprehensive Introduction R2R (RAG to Riches) is an advanced AI retrieval system supporting Retrieval Augmented Generation (RAG) functionality with production-ready features. Built on a containerized RESTful API, the system provides multimodal content parsing, hybrid search functionality...
Megrez-3B-Omni: an end-side multimodal understanding model supporting text, image, and audio multimodal understanding and analysis
综合介绍 Infini-Megrez是由无问芯穹（Infinigence AI）开发的边缘智能解决方案，旨在通过软硬件协同设计，实现高效的多模态理解和分析。该项目的核心是Megrez-3B模型，支持图...
GenEx: Generating explorable 360° 3D worlds from a single image (code progressively open-sourced)
综合介绍 GenEx 是一个先进的人工智能模型，能够从单张图像生成一个完全可探索的360° 3D世界。用户可以交互式地探索这个生成的世界。GenEx 推动了具象化人工智能在想象空间中的发展，并有潜力将...
RAGFlow: an open source RAG engine based on deep document understanding, providing efficient retrieval-enhanced generation workflows
Comprehensive Introduction RAGFlow is an open source Retrieval Augmented Generation (RAG) engine based on deep document understanding technology. It provides an efficient RAG workflow for organizations of all sizes, incorporating a large-scale language model (LLM) capable of delivering data in complex formats based on real...
NodeTool: a node orchestration-based workflow visualization client for AI models
General Introduction NodeTool is an innovative AI authoring platform designed to provide a simple, intuitive interface for AI enthusiasts, developers, data scientists and creatives. Whether you're an artist, developer, or beginner, NodeTool helps you quickly prototype creative...
Porkybank: AI-powered daily budget management for easy tracking of daily budgets
综合介绍 Porkybank 是一个开源的个人财务管理应用，旨在帮助用户轻松追踪每日预算。通过简单的公式（收入 - 支出）/ 天数 = 现金，用户可以直观地了解自己的财务状况。该项目托管在 GitHu...
CrewAI: A Multi-Roleplay Collaborative Intelligence Framework to Simplify Complex Tasks
Comprehensive Introduction CrewAI is an advanced framework designed to orchestrate collaboration between role-playing and autonomous AI agents. By facilitating collaborative intelligence, CrewAI enables agents to work together seamlessly to solve complex tasks. Whether you're building an intelligent assistant platform, automating customer service teams, or multi-agent...
Artab: show world famous paintings in a new tab opened in the browser, Chrome plugin
综合介绍 Artab 是一个浏览器扩展程序，旨在每次打开新标签页时展示世界上最伟大的艺术作品。该扩展程序适用于 Chrome、Edge 和 Firefox 浏览器。通过 Artab，用户可以在日常浏览...
Leffa: High-fidelity model virtual fitting and character pose adjustment, Meta open source controllable character image generation model
Comprehensive Introduction Leffa is a unified framework for generating controllable character images, enabling precise manipulation of character appearance (e.g., virtual fitting) and pose (e.g., pose transfer). The framework significantly reduces distortion of fine-grained details by directing the target query to focus on the correct reference key in the attention layer, with ...
MMAudio: generating synchronized sound effects and soundtracks for video footage, video-to-audio multimodal co-training tool
General Introduction MMAudio is an open-source project aiming to generate high-quality synchronized audio through joint multimodal training. Developed by Ho Kei Cheng et al. at the Chinese University of Hong Kong, the project's main function is to generate synchronized audio based on video and/or text input.MM...
H2O GPT: Flexible Configuration of Native AI Dialog and Document Processing Tools
综合介绍 H2O GPT 是一个开源项目，旨在提供私有化的聊天和文档处理功能。该项目基于 Apache 2.0 许可证，支持多种 GPT 模型，包括 LLaMa2、Mistral、Falcon 等。用...
OpenChat: Custom Chatbots for Fast Integration of Multiple Data Sources
综合介绍 OpenChat 是一个用户友好的聊天机器人控制台，旨在简化大语言模型（LLMs）的使用。通过提供两步设置流程，OpenChat 使用户能够轻松创建和管理多个自定义聊天机器人。该平台支持 G...
LocalGPT: Ensure data privacy by talking to multiple documents on local devices
General Introduction LocalGPT is an open source project designed to allow users to talk to documents on local devices and ensure data privacy. By using a variety of open source models, LocalGPT can process and understand document content without uploading data to the cloud. The project supports a variety of p...
English 
简体中文 
日本語 
한국어 
Español 
Deutsch 
Français 
Português do Brasil