Latest AI Resources

ROMA (Recursive-Open-Meta-Agent) is an open source meta-agent framework developed by Sentient AGI to efficiently solve complex problems through recursive task decomposition and parallel processing. Support for Python 3.12+, Docker and ...

GenFlow Super Hitchhiker is a general-purpose AI Agent launched by Baidu Literature Library, which allows users to autonomously disassemble tasks, call up Baidu Literature Library's 1.4 billion document libraries and online resources, and generate PPTs, reports, charts, posters, and other full-modal content in an extremely fast manner by simply typing in the natural language commands.

Nemotron 3 is a family of open source AI models released by NVIDIA in Nano, Super and Ultra sizes. It adopts the hybrid potential expert hybrid (latent MoE) architecture to significantly improve inference efficiency and reduce operating costs. Among them...

Hunyuan3D-Omni (Hybrid 3D-Omni) is an open source 3D asset generation framework by Tencent's Hybrid 3D team, which realizes accurate 3D model generation through multiple control signals. Based on Hunyuan3D 2.1 architecture, it introduces a unified control encoder that can handle point...

觅游（Meyo） 是美团基础研发 AI 创新产品团队推出的 AI 原生共生社区，目前已进入公测阶段。产品以"养虾"为核心隐喻，将 AI Agent 升级为拥有身份、MBTI 人格、社交关系与成长属性的...

Hybrid Motion1.0 (HY-Motion1.0) is Tencent Hybrid team open source text generated 3D action model , using 1 billion parameters Diffusion Transformer architecture , can be directly generated through natural language description of high-quality 3D character animation .

MoE-TTS is a speech synthesis framework introduced by KunlunWanwei, based on the Mixed Expert (MoE) architecture, which combines pre-trained Large Language Models (LLMs) with speech expert modules.MoE-TTS retains the powerful textual reasoning by freezing the textual module parameters and updating only the speech module parameters...

General Description eCommercePrompts offers over 2 million ready-to-use ChatGPT command recipes, these are built by command engineers and are designed to help eCommerce marketers build stores and promotions in less than an hour. The site covers a wide range of content types, including home...

LongCat-Video-Avatar is an advanced audio-driven video generation model built on LongCat-Video open-sourced by Meituan, focusing on generating hyper-realistic, lip-synchronized long videos with natural dynamics and consistent identity.

Ming-UniAudio is Ant Group's open source unified audio multimodal generation model that supports mixed input and output of text, audio, image and video. Using multi-scale Transformer and hybrid expert (MoE) architecture , through modality-aware routing mechanism to efficiently handle cross-modal ...

EXAONE 4.0 is a hybrid reasoning grand model from LG AI Research in Korea, blending general-purpose natural language processing and advanced reasoning capabilities. The model supports Korean, English, and Spanish, and is divided into a 32B professional version and a 1.2B end-side version. The professional version is suitable for legal, accounting...

Ring-1T-preview is an open source trillion-parameter big model of Ant Group, based on Ling 2.0 MoE architecture, pre-trained on 20T corpus, and trained in reasoning ability by self-developed reinforcement learning system ASystem. In natural language reasoning ...

MineContext is an active context-aware AI partner open-sourced by the ByteDance Viking team to help users efficiently manage massive amounts of information and improve the efficiency of knowledge work. Over the screenshot and content understanding technology, automatically record the user's daily operations (such as browsing the web, editing documents, etc.), support...

Kotae is an intelligent chatbot platform designed for small businesses and developed on ChatGPT technology. Supporting over 80 languages, the platform responds quickly and naturally to customer inquiries, generating accurate answers based on analyzing a user's website content, training documents, FAQs, etc. Kotae has no...

Mapify is an AI mind map generator tool launched by XMind team. It can quickly convert text, PDF, web pages, video, audio and other formats into structured mind maps, helping users efficiently extract and organize key information.

Doppl is an AI virtual fitting application launched by Google. After the user uploads a full body photo, the application supports the clothing picture or screenshot "wear" in the digital version of their own body, and can be converted from static pictures to AI-generated video, so that users can more truly feel the effect of clothing on the body.

Shangtang Ruying is an AI digital human video production platform launched by Shangtang Technology. Based on big model technology, the platform supports the creation of highly realistic digital human images and personalization, including facial features, clothing, hairstyles, and so on. The platform is equipped with sound cloning, video generation, automated data labeling, real-time interaction, and other functions...

DeepSeek-TUI 是美国独立开发者 Hunter Bown（GitHub: Hmbown）用 Rust 语言编写的开源终端 AI 编程智能体，专为 DeepSeek V4 系列模型（Pro/F...

Hailuo Video Agent is a zero-threshold AI video creation Agent launched by MiniMax, open Beta version. Through simple text input or image upload, high-quality creative videos can be generated in one click, covering a variety of application scenarios, such as broad...

Klic Studio (formerly Krillin AI) is an AI-based video translation, dubbing and voice cloning tool designed for video creators and content exporters. It supports one-click deployment of the whole process, which can complete the video from download to finished output in one click, and is adapted to Jittery, Xiaohongshu, B...

Qwen-Image-Edit is an all-purpose image editing model introduced by Ali Tongyi, built on the Qwen-Image architecture with 20 billion parameters. The model combines both semantic and appearance editing capabilities, and can perform low-level visual appearance editing on images (e.g., adding, deleting...

Qwen3-Omni is a fully modal AI model introduced by the Ali Tongyi team that can handle multiple data types such as text, images, audio and video, and supports text interaction in 119 languages with low latency and high controllability.

Staying ahead in the age of AI is an AI leadership guide from OpenAI that helps business leaders maintain a competitive edge in the age of AI. The guide points to the rapid growth of AI, with faster model releases, lower costs, and faster enterprise adoption...

FireRed-Image-Edit 是小红书 Super Intelligence 团队开源的通用图像编辑模型，基于扩散 Transformer 架构，在 GEdit、ImgEdit 等多个权威评测...

GPT-5 is the latest language model released by OpenAI with several upgrades. It is a unified intelligence system with a built-in real-time router that automatically switches between efficient and deep thinking modes according to the complexity of the problem, realizing fast response and accurate answers.GPT-5 has several versions, including the one for general...

General Introduction Hyperbolic AgentKit is an open source project designed to provide a template for running AI agents, combining blockchain and computing power. The project is based on Coinbase's CDP Agentkit modified and extended to support endpoints in...

RoboBrain 2.0 is an open source advanced embodied brain model that provides powerful perception, reasoning, and planning capabilities for robots.RoboBrain 2.0 consists of two versions, 7B and 32B, and utilizes a heterogeneous architecture that fuses visual encoders and language models to support...

General Description Stable Diffusion Web is an online platform for stable diffusion modeling that is free for users to use. The model is used to convert text into photorealistic images with an easy-to-use interface. By entering text prompts and clicking "Generate", users can quickly get...

Penguin Reading Companion is an AI reading assistant designed for primary and secondary school students by Tencent. Penguin Reading Companion relies on Tencent's hybrid big model and metamachine platform, combined with the Compulsory Education Language Curriculum Program and Curriculum Standards (2022 Edition), to provide students with personalized reading recommendations, multiple reading modes (focusing, reading aloud, listening...

微信ClawBot是微信官方推出的AI插件，核心定位是连接OpenClaw（开源AI智能体框架）与微信的消息通道。用户只需在部署OpenClaw的设备上执行一条命令完成安装，可通过微信聊天界面直接发送...

DeepSeek-V3.1-Terminus is an upgraded version of DeepSeek-V3.1, an artificial intelligence language model from the DeepSeek team. The model is optimized in terms of language consistency, code generation, and search capabilities to more accurately...

Elegant YOYA is a multimodal literate video platform launched by Zhongke Wenge, the platform is based on AI multimodal technology to empower the whole chain of video content creation. Users only need to input the theme requirements, the platform can quickly generate scripts, images, videos, and can complete intelligent editing, voice synthesis and character mouth drive and other operations, the output...

gpt-oss is a family of open source inference models from OpenAI that enable efficient, flexible, and easy-to-deploy AI solutions for developers. gpt-oss consists of two versions, gpt-oss-120B with 117 billion parameters and support for 8...

GLM-OCR 是智谱开源的 0.9B 轻量级专业 OCR 模型，在 OmniDocBench V1.5 以 94.6 分刷新 SOTA。兼顾“小体积”与“全场景”，扫描、手写、印章、多语混排、复杂表...

Gemini 2.5 Flash Image (codename nano banana) is a state-of-the-art image generation and editing model from Google that maintains the consistency of characters across different scenes and supports precise image editing through natural language, such as blurring backgrounds and removing stains.

Confucius3-Math is the first domestic open source reasoning model focused on mathematics education open-sourced by NetEaseYouDao. With 14 billion parameters, optimized for K-12 math education scenarios, it can run efficiently on a single consumer GPU (such as RTX 4090D), with an inference performance of about...

Nemotron Speech ASR是英伟达开源的实时语音识别模型，专为低延迟场景优化，支持24毫秒极速转录和多人并发对话。核心采用混合Mamba-Transformer MoE架构，通过固定状态缓...

General Introduction Curiosity is a project aimed at exploration and experimentation, primarily using the LangGraph and FastHTML technology stacks, with the goal of building a Perplexity AI-like search product. The core of the project is a simple...

Comprehensive Introduction "Why My Wife Yelling At Me" is a unique marriage relationship simulation website designed to help users understand their partner's emotional reactions and communication patterns through artificial intelligence. Users can input different scenarios and experience the reactions of their virtual partner, simulating real...

PilotDeck 是清华大学 THUNLP 实验室、面壁智能、OpenBMB 与 AI9Stars 联合研发并开源的 Agent 操作系统。以WorkSpace（工作舱）为核心架构，为每个任务提供独...

SkyReels-A3 is an audio-driven digital human creation tool from Kunlun World Wide Group. SkyReels-A3 is an audio-driven digital human creation tool, which can generate high-quality dynamic video content through simple inputs (e.g., portrait images and voice), make static photos "come alive", and replace lines for existing videos with new lip-syncs that the characters will automatically...

General Description AI Company Research Assistant is an artificial intelligence-based tool designed to automate the company research and lead generation process. By using technologies such as LangGraph and FireCrawl, the tool is able to generate detailed company research reports in just a few minutes. Users simply need to...

UltraEval-Audio是清华大学NLP实验室、OpenBMB和面壁智能联合开发的音频模型评测框架，最新版本为v1.1.0。专注于解决音频模型复现难、依赖冲突等问题，提供一键复现热门模型（如Vo...

ChatGPT Agent is a general-purpose AI Agent from OpenAI that combines multiple capabilities to autonomously accomplish complex tasks. Users only need to describe their needs in natural language, and the Agent can automatically select the appropriate tools, such as browsing the web, extracting information, running code...

General Introduction NeverEnds is a platform from Singapore that provides an unlimited creative space for all kinds of stories by converting text into video through an advanced model. Users are free to express creativity, convey emotions, as well as build their brand image on this platform. This platform is known for its unique...

Lumina-DiMOO is a new generation of unified model for multimodal generation and understanding launched by Shanghai Artificial Intelligence Laboratory (SAL) in conjunction with Huawei Rise at the World Artificial Intelligence Conference 2025. Based on the Rise AI basic hardware and software platform and the MindSpeed MM multimodal large model suite, it accomplishes...

Ring-2.5-1T 是蚂蚁集团百灵大模型团队开源的全球首个万亿参数混合线性架构思考模型，采用1:7 MLA与Lightning Linear Attention混合设计，激活参数量达63B。模型在...

Higress MCP is an innovative platform launched by Invest Today that supports the rapid transformation of traditional financial data APIs into modern MCP services.Higress MCP enables the transformation of REST APIs to MCP Server based on a simple configuration without the need to program...

Comprehensive introduction Wealth Digital People is a platform that integrates advanced AI technology, focusing on providing virtual image broadcasting and real-time interactive services. The platform utilizes self-developed speech recognition, speech synthesis, multimodal perception and document Q&A technologies to create realistic digital human doppelgangers for users, supporting video production, translation, teaching...

Genie 3 is a next-generation universal world model from Google DeepMind that enables the generation of highly dynamic and coherent virtual worlds in real time.Genie 3 simulates physical phenomena, natural ecosystems, and supports the creation of fantasy and historical scenarios. With text prompts, users can...

Magistral is an inference model from Mistral AI that focuses on transparent, multilingual and domain-specific reasoning capabilities. The model consists of an open source version (Magistral Small) and an enterprise version (Magistral Medium), the latter in...

JoyHallo is an open source AI digital human model from Jingdong, designed for Mandarin, supporting the conversion of audio into realistic speaking video.JoyHallo embeds audio features based on the wav2vec2 model, using a semi-decoupled structure to improve the accuracy of lip movement prediction, and supports the generation of English video...

General Introduction SagaLabs is an AI-powered translation tool and content platform designed for creators to help them localize and promote their stories to the global market. The platform supports translations in more than 200 languages, ensuring translation quality at the level of native language experts.SagaLabs is not only about...

Baichuan-M2 is an open source medical augmented large model launched by Baichuan Intelligence. It performs well in the medical field, especially in the HealthBench review with a score of 60.1, surpassing OpenAI's gpt-oss120b and many other open source models, becoming a global...

Qwen3Guard is a fine-tuned security protection model based on the Qwen3 base model, designed for security detection. It provides accurate security categorization of prompts and responses, provides risk levels, and supports English, Chinese, and multi-language environments.Qwen3Guard comes with two pro...

Cosmos 3 是 NVIDIA 开源的首个完全开放的物理 AI 全模态模型，采用创新的 Mixture-of-Transformers（MoT）架构，将视觉推理、物理世界生成与动作预测统一在单一系...

CRIC Depth Intelligence is the first AI intelligent body of Chinese real estate independently developed by CRIC, based on CRIC's 20 years of experience in the real estate industry and data accumulation and multimodal big model technology, which opens up the whole chain from data integration, intelligent analysis to content generation.

Clacky AI is a cloud-based development environment (CDE) for developers dedicated to creating L3-level AI programming assistants. The tool supports multiple programming languages, including Python, Node.js, Golang, Ruby, and Java, and can help developers quickly put...

Qwen3-ASR-Flash is Alibaba's latest high-precision speech recognition model, based on the Qwen3 base model, trained on massive multimodal data. It supports 11 languages and multiple accents, including Mandarin, Sichuan, Minnan, Wu, Cantonese and other dialects...

Kandinsky 5.0 is the latest video generation model series developed by Russian AI team, focusing on lightweight design and high performance performance. The first model in the series, Kandinsky 5.0 Video Lite, has only 2 billion parameters but surpasses similar 14B models, especially...

MiniMax Music 1.5 is an advanced AI music generation tool that supports generating up to 4 minutes of music based on users' natural language descriptions. The model supports a variety of music styles and mood customization, generating a natural and full vocal color, smooth transitions, richly layered arrangements...

Seed-OSS is a large family of language models open-sourced by the Byte Jump Seed team, focusing on long text and reasoning tasks. The model performs well in complex logical reasoning and multi-step reasoning, with high accuracy and efficient problem solving.Seed-OSS supports long text contexts up to 512K...

AntSK FileChunk is a free intelligent document slicing tool designed for RAG (Retrieval Augmented Generation) applications. Semantic as the core, the document will be intelligently sliced into semantically complete, coherent segments , support for multi-language , can dynamically adjust the size of the slice to ensure that the context of coherence.

PaperBanana是北大与谷歌团队联合开源的AI学术插图自动生成框架，专门解决科研人员绘制方法示意图和统计图表的痛点。框架通过五个智能体协作（检索、规划、造型、渲染和批评），实现从文本描述到Neu...

PaddleOCR-VL is Baidu's open source ultra-lightweight visual-language model, optimized for document parsing scenarios. The model contains only 0.9B parameters , through the fusion of dynamic high-resolution visual coder and lightweight ERNIE language model , while maintaining high accuracy and significantly reduce the computational overhead .

Qwen3-Coder-Flash is a high-performance programming model introduced by Ali Tongyi Thousand Questions team, which has excellent agent-based programming and tool invocation capabilities, and is good at handling complex programming tasks. The model supports 256K tokens of long context understanding, and can scale to 1M ...

Hunyuan-GameCraft is Tencent Hunyuan team open source interactive game video generation framework. Framework from a single picture and prompts to generate highly dynamic game video , support the user through the keyboard and mouse to control the video content in real time .

GLM-4.7-Flash是智谱开源的混合专家架构语言模型，参数规模为30B，激活参数量3B，上下文窗口达200K，最大输出令牌为128K。在编程能力上表现出色，SWE-bench验证集分数达59.2...

MuseSteamer is a large model for multimodal video generation launched by Baidu. The model can quickly generate high-quality dynamic video content based on text descriptions or images provided by the user, supporting a variety of clarity and functionality versions to meet the needs of different scenarios of creation.

Job AI resume is an efficient and convenient intelligent resume generation and optimization platform. Based on AI technology, the platform helps users quickly generate professional and personalized resumes. Users only need to enter basic information and experience, the platform can generate high-quality resume in a short time, providing 2800+ beautiful templates, covering a variety of positions.

Comprehensive Introduction HuixiangDou is a large-scale language model (LLM)-based group chat assistant designed to cope with group chat scenarios through a three-phase pre-processing, rejection and response process. It is able to answer user questions without causing information flooding. The project provides a complete web ...

Hunyuan 3D world model 1.0 (Hunyuan World 1.0) is Tencent's open source industry's first immersive roaming, interactive, simulation world generation model. The model integrates panoramic visual generation and hierarchical 3D reconstruction technology , support for text or image input to quickly generate 36...

Step-Audio-R1.1是阶跃星辰开源的全球首个原生语音推理模型，最新升级版本在权威评测榜单Artificial Analysis Speech Reasoning中以96.4%准确率登顶。模型...

ConverzAI is a recruiting platform based on AI technology that optimizes the hiring process. The platform is based on deploying virtual recruiters (AI Agents) to automatically collect candidate information from the Application Tracking System (ATS) and perform initial screening. The platform uses multiple channels such as phone, email, SMS...

Aholo Viewer是群核科技开源的3D高斯浏览器，支持在普通设备浏览器中流畅渲染10亿级高斯点的超大3D场景。核心技术采用chunk-based LOD树架构，相比李飞飞团队的Spark 2.0...

Mureka V7.5 is a state-of-the-art AI music generation model from Kunlun World Wide, focusing on Chinese songwriting. The model can accurately reproduce tones and playing techniques to generate natural, smooth and emotional vocals. Based on optimized automatic speech recognition (ASR) technology, Mureka V...

InternVLA-A1 is a large model of embodied operation open-sourced by Shanghai Artificial Intelligence Laboratory. It has the ability to understand, imagine, and execute the integration, and can accurately complete the task. The model fuses real and simulated operational data, and automates the construction of massive multimodal through large-scale virtual-real hybrid scene assets...

GLM-Experimental is an experimental large language model launched by Wisdom Spectrum AI, which has been launched on the Z.ai platform. The model has the ability to automatically generate PPTs. After users input topics or key points, the model can quickly generate clearly structured and standardized presentations, which will...

MedASR is a 105 million parameter medical speech recognition model open-sourced by Google, fine-tuned on a 5,000-hour desensitized clinical corpus, optimized for drug, dosage, and anatomical terminology, with a built-in 6-gram medical language model, and a word error rate of only 4.6 on the private radiology dataset RAD-DICT...

Vibe Kanban是开源的AI编程代理任务管理工具，专为同时使用多个AI编程助手（如Claude Code、Gemini CLI、Codex等）的开发者设计。通过看板形式统一管理任务进度，支持并行...

Leaning Wonderful Writer is an AI dissertation writing tool that provides an efficient and convenient solution for academic writing. The tool supports one-click generation of dissertation outlines, abstracts and first drafts of the body of the paper, which is applicable to different levels of academic needs such as undergraduate and master's degrees, and covers multi-disciplinary fields such as science and technology, liberal arts and social sciences.

FactSnap is a new generation AI information verification tool that helps users quickly verify the authenticity of web information. By integrating multiple models and search engines, it performs real-time verification of selected text while the user is browsing the web.

Clawith 是 DataElem 团队开源的"OpenClaw for Teams"——面向团队的多智能体协作平台。在保留 OpenClaw 灵魂与记忆能力的基础上，升级为 Aware 自主感知系...

Qwen3-VL-Reranker是阿里巴巴推出的多模态重排序模型，专门用于提升跨模态检索的精准度。与Qwen3-VL-Embedding协同工作：前者负责快速召回候选结果，后者通过深度跨模态交互（如...

FLUX.1 Kontext is an image generation and editing model from Black Forest Labs that provides context-aware image processing techniques. The model understands responses to text and image cues, performs tasks such as object modification, style conversion, and background replacement, while maintaining the corner...

SkyReels-V3是昆仑万维Skywork AI开源的多模态视频生成模型，被誉为视频生成领域的"全能型"标杆。模型基于"一核多支"的统一架构，在单一建模框架内集成三大核心能力：参考图像转视频、智能...

Skywork-SWE-32B is an open source 32B scale software engineering (SWE) autonomous code intelligences base model introduced by Kunlun World Wide Web. The model focuses on software engineering tasks, has powerful repository-level code repair capabilities, and can perform in complex scenarios with multi-round interactions and long text processing...

Shortbread is an AI comic generation tool that allows users to quickly create complete comics based on simple natural language descriptions.Shortbread doesn't require any drawing skills, just type in the text and AI quickly generates personalized comics with unique characters, scenes and emotions.Sho...

Kimi K2.5 是月之暗面发布的开源旗舰模型，采用 1T MoE 架构、激活 32B、上下文 256K token，原生支持图文视频多模态输入。在 Agent、代码、视觉理解三大基准均列开源第一...

gpt-realtime is an advanced speech model from OpenAI that supports direct audio processing to generate natural and smooth speech. The model supports multiple languages and styles, understands non-verbal cues such as laughter, and can switch between languages.

SAM Audio is an open source multimodal audio segmentation model introduced by Meta to accurately separate arbitrary target sounds from complex audio mixes. By combining textual, visual, and temporal dimensional cues, it enables flexible and efficient audio processing for tasks such as audio editing, denoising, sound extraction, and...

GLM-TTS is an open source industrial-grade speech synthesis system with powerful speech synthesis capabilities. Adopting a two-stage generation architecture: the first stage will be converted to text into speech token sequences, and the second stage will be converted into high-quality audio token sequences. The system supports only 3 seconds of voice samples to complete the sound...

LingBot-Depth是蚂蚁灵波科技开源的高精度空间感知模型，专门解决机器人在透明玻璃、反光物体等复杂场景中的深度识别难题。模型通过创新的"掩码深度建模"技术，在RGB图像基础上预测缺失的深度值

General Introduction Product Hunt Daily Chinese Hotlist is an automated tool based on GitHub Actions that generates a daily list of popular products on Product Hunt at regular intervals in the form of a Markdown file...

Protenix-v1是字节跳动ByteDance Seed团队推出的首个开源蛋白质结构预测模型，性能在严格对齐训练数据和模型规模后超越AlphaFold 3。模型具备显著的推理时扩展特性：通过增加采...

Uthana is a powerful AI 3D character animation generation platform. Users can input text descriptions, upload reference videos, or search motion libraries, and AI can quickly generate realistic animations and support adapting models with any bone structure. The platform is equipped with a variety of features such as style migration, API integration, customization...

You can use all vip features without membership, unzip and use, don't upgrade! Never upgrade! Don't upgrade! Link: https://pan.quark.cn/s/a120ee707f47 Extract Code: jHDN

LingBot-VA 是蚂蚁灵波开源的全球首个“自回归视频-动作世界模型”，把视频生成与机器人控制塞进同一 Transformer，每一步同时输出下一帧世界画面和对应动作，实现“边想边干”。

StoryMem是字节跳动与南洋理工大学联合开源的AI视频生成系统，专为解决多场景视频中角色和环境一致性问题。核心通过"视觉记忆库"技术，自动存储关键帧并在后续生成时参考，确保人物外貌、服装、场景元素...