Vision Parse is an open-source PDF document conversion tool that converts PDF files to Markdown format based on Vision LLMs (Vision Language Models). Vision Parse can intelligently identify and extract text and tables from PDFs, and...
The Language of Motion is a multimodal language model developed by Fei-Fei Li's team at Stanford University. It integrates verbal and nonverbal language from human actions. The model can process text, speech, and motion data to generate corresponding target...
Mathtutor on Groq is an AI-powered math tutoring tool built on the Groq architecture. Based on speech recognition, it allows users to ask math questions via voice. The tool has a powerful built-in mathematical engine that can perform real-time calculations and use LaTeX...
VE-Bench, recently released by the MMCAL research team at Peking University, is the first metric specifically designed for evaluating video editing quality. VE-Bench is designed to closely align with human perception, providing a more accurate assessment of video editing effectiveness...
EDTalk is an audio-driven lip-sync model jointly developed by Shanghai Jiao Tong University and NetEase, enabling independent control of lip movements, head posture, and emotional expressions. Simply upload an image, an audio clip, and a reference video to drive...
Video-analyzer is an open-source video analysis tool that combines Llama's 11B vision model and OpenAI's Whisper model to extract keyframes from videos, transcribe audio content, and generate detailed video descriptions. The tool supports full...
SPAR is a self-playing framework developed by the Zhipu team, which enhances the ability of large language models to follow instructions. The framework is based on the interaction of two roles: an internal generator and a perfecter. The generator executes instructions to produce responses, and the perfecter...
LowCodeEngine is an open-source low-code development framework from Alibaba. Based on simple drag-and-drop and configuration methods, it allows developers to quickly build complex system pages. The framework boasts powerful customization capabilities, an elegant development experience, and rich features...
Qwen-Agent is an open-source agent development framework based on the Qwen model, enabling developers to build intelligent agent applications using the Qwen model's instruction following, tool usage, planning, and memory capabilities. Qwen-Agent supports functions...
Mind GPT-3o is a multimodal end-to-end large-scale model launched by Li Auto, integrating voice, vision, and language understanding technologies to achieve efficient real-time interaction. Mind GPT-3o possesses memory, planning, tool usage, and expression capabilities, and can improve...
Langya is version 1.0 of a new generation of large-scale marine artificial intelligence model independently developed by the Institute of Oceanology, Chinese Academy of Sciences, focusing on the prediction of ocean state variables. The model combines advanced artificial intelligence algorithms with professional marine science knowledge, enabling...
DynamicControl is a new text-to-image (T2I) task framework developed by Tencent YouTu in collaboration with research institutions such as Nanyang Technological University and Zhejiang University. It integrates multimodal large language model (MLLM) inference capabilities. DynamicControl...
VideoPhy, jointly developed by UCLA and Google Research, is the first benchmark to evaluate the physics-based reasoning ability of video generation models. It measures whether the videos generated by the model adhere to real-world physics. The VideoPhy benchmark...
Large Motion Model (LMM) is a unified multimodal motion generation model jointly developed by the S-Lab at Nanyang Technological University, Singapore, and the SenseTime research team. LMM can handle various motion generation scenarios, including text-to-motion, music-to-dance, and more...
StereoCrafter is an innovative framework jointly developed by Tencent AI Lab and ARC Lab. It can convert monocular video (2D video) into stereoscopic 3D video, meeting the growing demand for 3D content. StereoCrafter is based on depth estimation,...
AgiBot World is an open-source dataset of millions of real-device data from Zhiyuan Robotics, aiming to promote the development of embodied intelligence. The dataset contains over eighty everyday skills, covering five core scenarios including home, dining, and industry. The data scale and quality are impressive...
GraphAgent is an intelligent graphical language assistant jointly developed by the University of Hong Kong and the Hong Kong University of Science and Technology (Guangzhou). It can process both structured (like graph connections) and unstructured (like text and visual information) data from the real world...
DeepSeek Engineer is an AI programming assistant that integrates the DeepSeek API. Its command-line interface allows users to read local files, create new files, and perform real-time differential editing on existing files. DeepSeek Engineer uses...
Psi R0 is Lingchu Intelligence's first end-to-end embodied model based on reinforcement learning. It supports complex operations through the collaboration of two dexterous hands, enabling the chaining and mixed training of multiple skills to generate an intelligent agent with reasoning capabilities, completing and closing loops...
FireCrawl is an open-source AI web crawler tool specifically designed for extracting web data and converting it into Markdown or other structured data. FireCrawl boasts powerful crawling capabilities, supports dynamic web page content processing, and provides intelligent crawling status updates...
VideoVAE+ (VideoVAE Plus) is an advanced cross-modal video variational autoencoder (Video VAE) developed by a team at the Hong Kong University of Science and Technology. It achieves significant compression of large-scale video movements by introducing a novel spatiotemporal separation compression mechanism and text guidance...
Languine is an AI-powered translation tool that helps developers streamline the application development process. Based on intelligent detection, AI translation, automated workflows, and a developer-friendly design, Languine makes translation management...
Figma-Low-Code is an open-source project based on the Luisa framework, allowing developers to directly convert Figma designs into Vue.js applications. It significantly reduces handover time between designers and developers, and reduces front-end coding workload...
AI Dev Gallery is an open-source AI toolkit and sample library from Microsoft, integrated into Visual Studio, helping Windows developers easily integrate edge AI functionality. AI Dev Gallery provides over 25 interactive examples,...