Crawl4AI is an asynchronous web crawling framework developed in Python, designed specifically for large-scale language modeling (LLMs) and artificial intelligence (AI) applications, simplifying the web crawling and data extraction process. Based on an asynchronous architecture, it efficiently processes...
Wren AI is an open-source text-to-SQL solution based on natural language processing technology. It allows users to execute database queries by asking questions in natural language, eliminating the need to write complex SQL code. It supports various databases and...
Podcastfy is an open-source Python package that converts web content, PDF documents, and text into multilingual audio conversations. This tool utilizes advanced generative artificial intelligence (GenAI) technology, similar to...
Gradio is an open-source Python library that simplifies the demonstration and sharing of machine learning models. It allows developers to quickly create user-friendly web interfaces with simple code, enabling anyone, anywhere, to easily use machine learning...
LosslessCut is an open-source video editing tool for quickly and losslessly cutting and merging video and audio files. Built on the powerful FFmpeg library, it provides an intuitive graphical user interface, allowing users to easily cut and merge...
Future You is an AI-powered conversational chat product developed by MIT, allowing users to interact with a virtual avatar at age 60. Future You is based on advanced AI technology, combined with the user's current life goals and...
PixWizard is a versatile image-to-image visual assistant that performs tasks such as image generation, editing, and translation based on natural language commands. The system integrates various visual tasks through a unified image-to-text to image generation framework...
Screenpipe is an open-source software that uses AI technology to monitor a user's computer screen and microphone around the clock. By capturing screen activity and audio, it uses AI for intelligent analysis to improve work efficiency and protect data privacy. Screenpipe...
Pyramid-Flow is an advanced video generation model jointly developed by researchers from Peking University, Kuaishou Technology, and Beijing University of Posts and Telecommunications. The model can generate videos up to 10 seconds long with a resolution of up to 1280x768 and a frame rate of 2... based on text prompts.
Aria, developed by the Rhymes AI team, is the world's first open-source, native multimodal hybrid expert (MoE) model capable of understanding and processing various input modalities, including text, code, images, and video. The model demonstrates strong performance on multimodal and language tasks...
Swarm is an experimental framework developed by OpenAI, designed to build, orchestrate, and deploy multi-agent systems. Based on lightweight agents and handoff mechanisms, it simplifies the coordination and execution processes between agents, enabling more precise control...
Baidu AI Cloud's Yijian is a visual large-scale model platform launched by Baidu AI Cloud. It aims to democratize professional-grade visual AI applications based on leading visual large-scale model technology and rich scenario-based algorithm solutions. The platform connects the entire process from model generation...
libcom is an image compositing toolkit developed by the Shanghai Jiao Tong University (BCMI) laboratory. It aims to solve inconsistencies between foreground and background, such as mismatches in appearance, geometry, and semantics, to generate realistic composite images...
F5-TTS is a high-performance text-to-speech (TTS) system open-sourced by Shanghai Jiao Tong University. It is based on a non-autoregressive generation method using stream matching, combined with Diffusion Transformer (DiT) technology. The system can...
CursorCore is a programming aid framework based on Large Language Models (LLMs) that uses conversational interaction to help programmers write and modify code more efficiently. The framework integrates various information from the programming process, including code history, ...
MM1.5 is a multimodal large-scale language model released by Apple, designed to enhance rich image understanding, visual reference and localization, and multi-image reasoning capabilities. The model is based on a data-centric training approach, utilizing large-scale pre-training...
Surya is a powerful open-source OCR (Optical Character Recognition) toolkit specifically designed for document recognition, supporting over 90 languages. Surya can accurately recognize text in documents and analyze the reading order of the text...
Illuminate is a Google project that transforms academic papers into AI-generated audio discussions. Based on Google's powerful language model Gemini, the project converts paper content into engaging dialogues, making...
Loong is a novel long-form video generation model jointly developed by the University of Hong Kong and ByteDance. It can generate minute-long videos with consistent appearance, rich dynamics, and natural scene transitions. The model is based on an autoregressive large language model (LLM)...
ScriptViz is a script visualization tool developed by researchers at Stanford University. Based on the large film database MovieNet, it retrieves matching movie scenes from the script text and dialogue, and visualizes the screenwriter's written descriptions...
MLE-bench is a benchmarking tool launched by OpenAI, designed to measure the performance of AI agents in machine learning engineering tasks. The test includes 75 competition tasks from Kaggle, covering natural language processing...
GTSinger is a large-scale, open-source, high-quality singing dataset developed by a research team at Zhejiang University, designed to support diverse singing tasks. GTSinger contains 80.59 hours of professionally recorded vocals, covering nine different languages...
LightRAG is a retrieval-augmented generation (RAG) system developed by a research team at the University of Hong Kong. Based on an integrated graph-structured index and a two-layer retrieval mechanism, it enhances the performance of large language models...
Gen2Act is a robot manipulation strategy jointly developed by Google, Carnegie Mellon University, and Stanford University. It generates human-like videos based on motion information from predictive network data and uses these videos to guide robots in performing new tasks...