xGen-MM is an open-source multimodal AI model from Salesforce. It has the ability to process interleaved data and simultaneously understand and generate multiple data types, including text and images. xGen-MM learns from a large amount of image and text data...
Phi-3.5 is Microsoft's next-generation AI model series, comprising three versions: Phi-3.5-mini-instruct, Phi-3.5-MoE-instruct, and Phi-3.5-vision-instruct, targeting lightweight inference, hybrid...
TrackGo is an advanced, controllable AI video generation technology that provides users with a flexible and precise mechanism to control the movement of objects in videos using free-form masks and arrows. The core of the technology is the TrackAdapter...
GPT Pilot is an AI programming tool that helps build applications from scratch by mimicking the workflow of human developers. It allows users to write code, debug programs, discuss issues with users, and request code reviews—it's a true...
GaussianEditor is a 3D editing algorithm based on Gaussian Splatting, allowing users to quickly and accurately modify 3D scenes. Through Gaussian semantic tracking and hierarchical Gaussian Splatting, it provides a high degree of editing...
Seed-ASR is an ASR (Automatic Speech Recognition) model developed by ByteDance based on a Large Language Model (LLM). It was trained on over 20 million hours of speech data and nearly 900,000 hours of paired ASR data, supporting Mandarin and 13...
mPLUG-Owl3 is an advanced, general-purpose, multimodal large model launched by Alibaba, designed specifically for understanding and processing multi-image and long video files. While maintaining accuracy, it significantly improves inference efficiency, capable of analyzing a 2-hour movie in 4 seconds.
Moffee is an open-source Markdown to PowerPoint tool that allows users to quickly create professional-looking presentations using Markdown syntax. Moffee automatically handles layout, pagination, and styles, and supports real-time preview and export to PDF or HTML format.
Lingo is the first end-to-end large-scale voice model in China launched by Westlake Xinchen. Technically, it has the ability to interrupt in real time, control commands in real time, be super human-like, and sing and speak. It has better Chinese voice effect than GPT-4o.
The ShanHai Big Model is a multimodal AI big model launched by Unisound, possessing rich knowledge reserves and multimodal interaction capabilities. The ShanHai Big Model can interact with users in real time through various formats such as text, audio, and images, providing information...
AIEditor is a next-generation rich text editor designed for AI. Developed based on Web Components, AIEditor supports virtually any front-end framework, including Layui, Vue, React, Angular, and Svelte. AIEditor is compatible with...
Omages is an open-source 3D model generation project that uses image diffusion technology to encode the geometric and texture information of 3D shapes into 64x64 pixel 2D images, thereby simplifying the 3D modeling process.
Puyu Lingbi IXC-2.5 is a new generation of multimodal large-scale model developed by the Shanghai Artificial Intelligence Laboratory. It features a large-scale language model backend of 7 bytes. It can handle long contexts of up to 96KB and supports ultra-high-resolution images and fine-grained video...
GLM-4-Flash is the first free large-model API launched by Zhipu AI. GLM-4-Flash not only supports multi-turn dialogue and multi-language processing, but also features advanced functions such as web browsing and code execution. Developers and enterprises can use Zhipu AI to...
AgentGen is an AI project framework jointly launched by the University of Hong Kong and Microsoft. It significantly improves the planning capabilities of Large Language Models (LLMs) by automatically generating diverse environments and tasks. AgentGen utilizes the BI-EVOL method to create challenging...
MotionGen is a 3D motion generation model developed by MotionGen Technology. It combines large models, physics simulation, and reinforcement learning algorithms, allowing users to quickly generate realistic and fluid 3D motions using only simple text commands. MotionGen...
NewOne Video's Big Model is an AI video creation big model launched by NewOne Technology. It utilizes self-developed AI algorithms and deep learning technology to achieve one-click creation from script to finished product. It features script generation, emotional voice synthesis, 3D element creation, and...
GameNGen, Google's first AI game engine, can generate realistic DOOM game visuals in real-time at 20 frames per second, making it indistinguishable from reality for 60% of players. GameNGen suggests that game development may no longer require traditional methods...
GLM-4-Plus is the latest high-intelligence flagship model from Zhipu AI, achieving breakthroughs in language understanding and long text processing. It employs innovative technologies such as PPO to significantly enhance reasoning and instruction following capabilities. GLM-4-Plus excels in several key areas...
Shusheng Puyu is an open-source AI model developed by the Shanghai Artificial Intelligence Laboratory, boasting exceptional reasoning capabilities and the ability to process extremely long texts. Shusheng Puyu supports text input of up to one million words and can autonomously perform web searches...
CogView-3-Plus is the latest AI text-to-image generation model from Zhipu AI. It uses the Transformer architecture instead of the traditional UNet and optimizes noise planning in the diffusion model. CogView-3-Plus performs exceptionally well in image generation, capable of...
GLM-4V-Plus is the latest multimodal AI model from Zhipu AI, focusing on image and video understanding. GLM-4V-Plus can not only accurately analyze static images, but also possess temporal awareness and understanding capabilities for dynamic video content, capturing...
MetaHuman-Stream is a cutting-edge real-time interactive streaming AI digital human technology that integrates multiple advanced models such as ERNerf, MuseTalk, and Wav2lip. It supports voice cloning and deep learning algorithms to ensure smooth and natural dialogue...
Qwen2-VL is an open-source visual multimodal AI model from Alibaba DAMO Academy, possessing advanced image and video understanding capabilities. Qwen2-VL supports multiple languages, can process images of varying resolutions and aspect ratios, and can analyze dynamic videos in real time...