Latest AI Resources

LLaVA-OneVision-1.5 is an open-source multimodal model by the EvolvingLMMS-Lab team, using 8B parameter scale, through a compact three-phase training process (language-image alignment, conceptual equalization and knowledge injection, and instruction fine-tuning) on 128 A800...

Paper2Video is an open-source presentation video project for automatic generation of academic papers by Show Lab at National University of Singapore. Using the PaperTalker multi-intelligence framework, papers are transformed into full presentation videos containing slides, subtitles, voiceover and speaker avatar...

NeuTTS Air is open source lightweight speech synthesis model, developed by Neuphonic team, which can run in real time on local devices (e.g. cell phones, laptops, Raspberry Pi) without relying on the cloud. Using 0.5B parameter Qwen architecture and self-developed NeuCodec codec...

KAT-Dev-72B-Exp is an open-source programming-specific large language model launched by the Racer team, optimized based on reinforcement learning technology, which achieved an accuracy rate of 74.6% in the SWE-Bench Verified benchmark test, the best performance of any open-source model at present. The model uses innovative...

Jamba Reasoning 3B is a lightweight inference model open-sourced by Israeli AI startup AI21 Labs with strong performance and potential for a wide range of applications. It utilizes a hybrid SSM-Transformer architecture that combines Trans...

Agentic AI is the newest course on intelligent bodies launched by Ernest Ng.The course focuses on the design and construction of intelligent bodies, covering the four major design patterns of reflection, tool use, planning, and multi-intelligent body collaboration. Learners will master how to make intelligent bodies check outputs, autonomously adjust through theoretical explanations and code practice...

OpenAgents is the open source project that creates a network of AI agents and facilitates open collaboration between agents. A basic network infrastructure is provided to enable AI agents to seamlessly connect and collaborate. Users can quickly start their own agent network, extend functionality through a modular architecture, support...

Androidify is Google's open source project to help developers learn how to build AI-driven apps on Android. The project uses Google's latest technologies such as Jetpack Compose, Gemini API (via Fire...

Ling-1T is a trillion-parameter general-purpose language model open-sourced by Ant Group, which belongs to the flagship product of the Ling 2.0 series of Bering's large models. The model adopts a highly efficient MoE architecture, supports 128K context windows, and surpasses GPT in 7 benchmarks including code generation, mathematical reasoning, and logic test...

EchoCare is a large model of ultrasound base developed by the Center for Artificial Intelligence and Robotics Innovation (CAIR) at the Hong Kong Institute of Innovation and Research of the Chinese Academy of Sciences (CAS), trained based on the world's largest ultrasound image dataset (more than 4.5 million images), covering multi-center, multi-region, multi-ethnicity, and more than 50 individuals...

Code2Video is innovative open source project that automatically converts code snippets into high quality video content (mp4 format). The project through a unique code-centric paradigm , the use of carbon-now-cli tools to generate code into beautiful images , the use of ffmpeg will be these ...

SceneGen is an open source method for generating 3D scenes from a single image at Shanghai Jiao Tong University. From a single scene image and a target resource mask, a complete scene containing multiple 3D resources is efficiently generated, including the geometric structure of the resources, texture and relative spatial location.

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 ...

AIMangaStudio is a free AI manga creation tool that provides creators with a complete manga creation pipeline, including plot generation, sub-scene design, character setting and other functions, which can simplify the production process from script to manga page. It supports natural language generation of comic scripts, including plot, dialog...

FireRedChat is an open source full-duplex voice interaction system for Xiaohongshu with real-time bidirectional dialog capabilities and support for controlled interruptions. Adopts a modular design , including transcription control module , interaction module and dialogue manager , etc., supports cascade and semi-cascade architecture , can be flexibly deployed .

Logics-Parsing is an open source Ali end-to-end document parsing model , based on Qwen2.5-VL-7B. Optimize document layout analysis and reading order inference through reinforcement learning , PDF images can be converted to structured HTML output to support a variety of content ...

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 ...

RoboBrain-X0 is the world's first open source embodied model that supports zero-sample cross-ontology generalization open-sourced by Wisdom Source Research Institute, which is of great industrial significance. It can drive multiple real robots of different configurations to complete basic operation tasks without fine-tuning, and after a small amount of sample fine-tuning, it demonstrates the ability to replicate ...

Lynx is a high-fidelity personalized video generation model open-sourced by ByteDance that can generate identity-consistent videos with only a single portrait photo. Built on the diffusion Transformer (DiT) base model , the introduction of ID-adapter and Ref-adapte...

Claude Sonnet 4.5 is an artificial intelligence model from Anthropic designed for programming, computer operations, and complex task automation. The model excels in code generation, long-duration task processing, reasoning, and mathematical computation, supporting everything from initial planning...

DeepSeek-V3.2-Exp is a DeepSeek open source experimental AI model that significantly improves the efficiency of long text processing by introducing the DeepSeek Sparse Attention (DSA) mechanism. The model is based on DeepSeek...

HunyuanImage 3.0 (HunyuanImage 3.0) is a native multimodal image generation model released and open-sourced by Tencent. The model parameter size of 80B, is currently the best evaluation results, the largest number of parameters of the open source image generation model. Hybrid Image 3.0 supports real-time image generation, users can side...

Hunyuan3D-Part (Hybrid 3D-Part) is a 3D generation model released and open-sourced by Tencent. Composed of P3 - SAM and X - Part, it realizes high-precision and controllable component-based 3D generation for the first time, and supports 50 + components to be generated automatically. Users can first use...

AudioFly is KDDI open source AI model for text to generate sound effects. Based on the potential diffusion model architecture, with 1 billion parameters, trained on large-scale, diverse audio text datasets, covering AudioSet, AudioCaps, TUT and other public datasets and internal...

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...

FLM-Audio is a native full-duplex audio dialog grand model released by Beijing Zhiyuan Artificial Intelligence Research Institute in conjunction with Spin Matrix and Nanyang Technological University of Singapore, supporting both Chinese and English. Adopting native full-duplex architecture, it can merge listening, speaking and monologue at each time step...

CWM (Code World Model) is a 32-billion-parameter open-source world language model released by the Meta FAIR team, designed for code generation and reasoning. Introducing the concept of "world model", it can simulate the code execution process, predict the variable state changes, and advance...

Neovate Code is an open source intelligent programming assistant from Ant Group's Alipay Experience Technology Department, which improves development efficiency through artificial intelligence technology. With conversational development features, developers can describe the requirements through natural language, Neovate Code can understand and generate the corresponding generation...

Audio2Face is NVIDIA's open source AI tool capable of transforming audio input into realistic 3D facial animation. By analyzing speech features in the audio, such as phonemes and intonation, it generates precise lip synchronization and subtle emotional expressions to give vivid human expressions to virtual characters.

Qwen3-VL is an open source multimodal visual language large model by AliCloud Tongyi Qianqian team, the number of references reaches 235 billion, and the model file is about 471GB.Containing instruction version and thinking version, it adopts enhanced MRope interleaved layout, DeepStack and other technologies, which can effectively utilize the visual transform...

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...

Qwen3-TTS-Flash is an advanced speech synthesis model introduced by Ali Tongyi, supporting 17 tones and 10 languages, covering Mandarin, English, dialects, etc. It has excellent stability and high expressiveness of Chinese and English speech, and the model can automatically adjust the tone of voice to make it more vivid.

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.

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...

Granite-Docling-258M is an ultra-compact open source visual language model from IBM designed for efficient document conversion. The model converts documents into machine-readable formats while leaving layout, tables, formulas, and other elements intact.

Lucy Edit is an open source AI video editing tool developed by Decart AI. Allows users to edit video through simple natural language descriptions, such as "change the character into a polar bear" or "turn the scene into a 2D cartoon style", without the need for complex fine-tuning or the use of masks ...

LongCat-Flash-Thinking is a highly efficient reasoning model released by the LongCat team at Mission LongCat that has become more powerful and specialized while maintaining the extreme speed of LongCat-Flash-Chat. The model is based on logic, math, code, intelligence...

Ling-V2 is a family of large-scale language models based on the MoE architecture introduced by the Ant-Belling team. The first version, Ling-mini-2.0, has 16 billion total parameters, with only 1.4 billion parameters activated per input token.

Kronos is the first K-line chart base model for financial markets jointly open-sourced by Tsinghua University and Microsoft Research Asia. It analyzes K-line data of stocks, cryptocurrencies and other assets, including opening, high, low, closing and volume, to predict future price movements.

Wan2.2-Animate is an open source action generation model , support for action imitation and role-playing mode . Users only need to input a character picture and a reference video , the model can migrate the video character's movements and expressions to the picture character , giving the picture character dynamic expression ...

Xiaomi-MiMo-Audio is Xiaomi's open source 7-billion-parameter end-to-end speech macromodel with powerful features such as multi-language dialog, speech continuation, less-sample generalization, and audio understanding, which is able to reach the SOTA level in speech intelligence and audio understanding benchmarks, surpassing Google Gemi...

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...

VoxCPM is a speech generation model jointly open-sourced by Facade Intelligence and Shenzhen International Graduate School of Tsinghua University.VoxCPM adopts an end-to-end diffusion autoregressive architecture to generate continuous speech representations directly from text, breaking through the limitations of traditional discrete disambiguation. Through hierarchical language modeling and finite state quantization...

InternVLA-N1 is an open source end-to-end dual-system navigation macromodel from Shanghai Artificial Intelligence Laboratory. Using a dual-system architecture, System 2 is responsible for understanding linguistic commands and planning long-range paths, while System 1 focuses on high-frequency response and agile obstacle avoidance. The model is trained entirely based on synthetic data through large-scale digital ...

WebWeaver is a new dual-intelligence body framework introduced by Alibaba Tongyi team, which is mainly used in open deep research, and can simulate the human research process, which is divided into two intelligences: planning and writing.

The MCP Registry is a centralized platform from GitHub that helps developers discover and install MCP servers more easily.The MCP Registry is here to help developers quickly find the AI tools they need in one place, greatly simplifying...

VLAC is an open source embodied reward macromodel from Shanghai Artificial Intelligence Laboratory. Based on InternVL multimodal macromodel, it integrates Internet video data and robot operation data to provide process reward and task completion estimation for robot reinforcement learning in the real world.VLAC can effectively ...

Tongyi DeepResearch (Tongyi DeepResearch) is an open source intelligent body launched by Alibaba, designed for deep information retrieval and complex task reasoning, with 30 billion parameters, supporting multiple reasoning modes, including ReAct mode and deep mode...

InternVLA-M1 is an open-source embodied operating "brain" of Shanghai Artificial Intelligence Laboratory, which is a large model of two-system operation oriented to instruction following. It builds a complete closed loop covering "think-act-learn" and is responsible for high-level spatial reasoning and task planning. The model adopts a two-phase training cur...

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...

Fundamentals of Large Models provides an in-depth analysis of the core technologies and practical paths of Large Language Models (LLMs). Starting from the fundamental theory of language modeling, it systematically explains the principles of model design based on statistics, recurrent neural networks (RNN), and Transformer architecture, focusing on the three major big language model...

PromptEnhancer is an open source prompt word enhancement tool from Tencent's Mixed Meta team to improve the generation of text-to-image (Text-to-Image, T2I) models. Through the chain of reasoning (Chain-of-Thought, CoT) approach to the use of ...

LLaSO is an open source speech model launched by Beijing Depth Logic Intelligence Technology Co. Ltd, which solves the problems of data dispersion and insufficient task coverage in the field of large-scale speech language modeling by integrating speech and text data and providing alignment datasets, command fine-tuning datasets and evaluation benchmarks.

Hybrid 3D 3.0 is an advanced 3D generation model launched by Tencent, based on 3D-DiT hierarchical sculpting technology, with a geometric resolution of up to 1536³, capable of generating ultra-high-definition, detail-rich 3D models, and excelling in character modeling, with the ability to accurately shape the five senses and body shape.

UnifoLM-WMA-0 is an open source world model-action architecture across multiple classes of robot ontologies by Yu Shu Technology, designed for general robot learning. Composed of a world model and an action architecture, the world model understands the physical laws of robot-environment interaction, and the action architecture is responsible for specific...

InfiniteTalk is an audio-driven video generation tool developed by the MeiGen-AI team that generates talking videos of unlimited length based on the input audio. The core advantage lies in the precise lip synchronization technology, which can perfectly match the audio with the character's mouth shape to generate natural and smooth...

Mini-o3 is an open source model jointly launched by ByteDance and the University of Hong Kong, focusing on solving complex visual search problems. The model has a powerful multi-round interactive reasoning capability, and can locate the target through deep exploration and trial-and-error.

GPT-5-Codex is a powerful programming optimization model from OpenAI, further enhanced by GPT-5 and designed for software engineers. The model generates high-quality code quickly, supports multiple programming languages, and optimizes existing code to improve performance.

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 ...

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...

Hyprnote is an open source, local-first AI meeting note-taking tool designed for professionals to protect user privacy and improve meeting efficiency. Adopting the "local first" principle, all data storage and processing is done on the user's local device to ensure data security and support offline operation.

MobileLLM-R1 is Meta's open source series of efficient inference models designed for mathematical, programming and scientific reasoning. It contains a base model and a final model, with 140 million, 360 million and 950 million parameter versions, respectively. The models are not generic chat models and are supervised fine-tuned (SFT...

ERNIE-4.5-21B-A3B-Thinking is Baidu's open source large-scale language model focused on reasoning tasks. Using the Mixed Expert (MoE) architecture , the total number of references to 21 billion , each token activates 3 billion parameters to support 128K long context window ...

MobiAgent is an open source mobile intelligent body toolchain from IPADS Lab of Shanghai Jiaotong University, which helps users to build their own mobile intelligent assistants. By recording the user's operation trajectory and generating high-quality data, it trains an intelligent body that can understand natural language commands. Core features include efficient...

ZipVoice is a series of speech synthesis (TTS) models based on the Flow Matching architecture released by Xiaomi, including ZipVoice (zero-sample single-speaker speech synthesis model) and ZipVoice-Dialog (zero-sample conversational speech synthesis...

PP-OCRv5 is the latest generation of text recognition AI model released by Baidu. With a lightweight design and a reference count of only 0.07B, it is suitable for efficient operation on CPU and edge devices, and can process more than 370 characters per second. The model supports Simplified Chinese, Traditional Chinese, English, Japanese and Pinyin...

Youtu-GraphRAG is an open source graph retrieval augmentation generation framework from Tencent's Youtu Labs to help large language models handle complex Q&A tasks more accurately. By constructing a four-layer knowledge tree, the knowledge is disassembled into four levels of attributes, relationships, keywords and communities to realize the self-directed performance of cross-domain knowledge...

Stand-In is a lightweight, plug-and-play identity-preserving video generation framework from Tencent's WeChat Vision team. Focusing on preserving specific identity features in video generation, it only needs to train the additional parameters of the base model 1%, and can achieve excellent results in face similarity and naturalness.

IndexTTS2 is a new free text-to-speech (TTS) model open-sourced by the B station voice team, which realizes a major breakthrough in emotional expression and duration control, the first autoregressive TTS model that supports precise duration control. Supports zero-sample voice cloning, only one audio file can accurately copy the sound...

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...

HuMo is a multi-modal video generation framework jointly open-sourced by Tsinghua University and ByteDance Intelligent Creation Lab, focusing on human-centered video generation. It can generate high-quality, fine-grained and controllable human videos from a variety of modal inputs such as text, images and audio.HuMo supports a powerful text cue-following capability...

AnyI2V is an image animation generation framework jointly launched by Fudan University, Alibaba Dharma Institute and others, which supports the conversion of static conditional images (e.g., grids, point clouds, etc.) into dynamic videos without the need for complex training processes and large amounts of data.

SRPO (Semantic Relative Preference Optimization) is a text-to-image generation model introduced by Tencent Hybrid, which optimizes the reward mechanism through text conditioned signals to achieve online adjustment of rewards and reduce offline fine-tuning dependency.

Qwen3-Next is a new generation of hybrid architecture big model open source by Ali Tongyi, combining Gated DeltaNet and Gated Attention technology, good at dealing with long text, fast inference and saving computing resources.

Wenxin Big Model X1.1 is a deep thinking model launched by Baidu, based on a hybrid reinforcement learning framework that focuses on improving language understanding and generation. The model excels in handling complex questions, following instructions and simulating the behavior of intelligences, and can accurately provide knowledgeable answers and high-quality text content.

HunyuanImage 2.1 is Tencent's open source graphic model, designed for high-quality image generation. The model supports native 2K resolution, can accurately render complex scenes and details, so that the character's expression and movement can be vividly reproduced.

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.

UnifiedTTS is a one-stop platform for text-to-speech (TTS) services. It supports multiple languages, including Chinese, English, Japanese and Korean, to meet the needs of global business. Through a unified API interface, it integrates many mainstream TTS services, including Micro...

MiniCPM 4.1 is an ultra-efficient end-side large language model introduced by Facade Intelligence. With InfLLM v2 sparse attention architecture, each lexeme only needs to calculate the relevance to less than 5% lexemes, which significantly reduces the processing overhead of long text. In a 128K long text scenario...

WeKnora is Tencent WeChat team open source based on the Large Language Model (LLM) document understanding and semantic retrieval framework , designed for the structure of complex, heterogeneous document content scenarios and designed to use a modularized architecture , integration of multimodal preprocessing , semantic vector indexing , intelligent recall and large model generative reasoning ...

XTuner V1 is a new generation of large model training engine open-sourced by Shanghai Artificial Intelligence Laboratory (SAL), designed for ultra-large scale sparse Mixed Expert (MoE) model training. Developed based on PyTorch FSDP, it achieves high performance through multi-dimensional optimization of memory, communication and load ...

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...

LangChain for LLM Application Development is an online course presented by DeepLearning.AI, featuring LangChain founder Harrison Chase and Andrew Ng.

Transformer LLMs work on the principle that DeepLearning.AI and Jay Alammar and Maarten Grootend, authors of Hands-On Large Language Models...

Seedream 4.0 is an advanced image generation and editing tool launched by ByteDance, centered on the integration of generation and editing, with powerful features such as precise command editing, high feature retention, and deep intent understanding.

rStar2-Agent is an advanced AI mathematical reasoning model open-sourced by Microsoft that demonstrates strong mathematical problem solving capabilities by achieving an accuracy of 80.61 TP3T in the AIME24 test. The model is equipped with scientific reasoning capabilities, achieving in the GPQA-Diamond benchmark...

Qwen3-Max-Preview is the latest flagship large language model released by Tongyi Qianwen. It is the model with the largest number of parameters in the Qwen3 family, with a parameter size of over 1 trillion. The model has significant improvements in inference, instruction following, multi-language support and long-tail knowledge coverage...

OneCAT is a new unified multimodal model launched by Meituan in conjunction with Shanghai Jiaotong University, which adopts a pure decoder architecture and can seamlessly integrate multimodal comprehension, text-to-image generation and image editing functions. The model abandons the design of traditional multimodal models that rely on external visual coders and disambiguators through modality-specific...

Claudable is an open source web application builder based on Next.js that combines the advanced AI agent capabilities of Claude Code and Cursor CLI with Lovable's simple and intuitive application building experience....

FineVision is Hugging Face's open source visual language dataset for training advanced visual language models. It contains 17.3 million images, 24.3 million samples, 88.9 million rounds of dialog, and 9.5 billion answer tokens. The dataset aggregates...

InfinityHuman is a commercial-grade long time-series audio-driven character video generation model jointly launched by ByteDance and Zhejiang University. The model is audio-driven and can generate high-resolution, long duration and visually consistent character videos.

Kimi K2-0905 is an advanced AI model from Dark Side of the Moon Technologies Ltd. that excels in programming assistance, generates code efficiently, and supports the generation of neat and standardized code in front-end development. The model context length is extended to 256K to handle complex tasks.

HunyuanWorld-Voyager (Hunyuan Voyager for short) is the industry's first ultra-long roaming world model released by Tencent that supports native 3D reconstruction. It is a novel video diffusion framework that generates a 3D point cloud sequence of user-defined camera paths from a single image, supporting...

Hunyuan-MT-7B is a lightweight translation model introduced by Tencent's Mixed Meta Team, with 7 billion references, supporting the mutual translation of 33 languages and 5 folk-Chinese languages/dialects, including Cantonese, Uyghur, and Tibetan. In the International Association for Computational Linguistics (ACL) WMT2025 competition...

Step-Audio 2 mini is an open source end-to-end speech grand model of Step-Audio. It breaks through the traditional speech model structure and adopts the true end-to-end multimodal architecture, which directly transforms the original audio input into speech response output with lower latency, and understands paralinguistic information and non-vocal signals.

MobileCLIP2 is an upgraded version of MobileCLIP, an efficient end-side multimodal model introduced by Apple researchers. It is optimized in terms of multimodal reinforcement training by training better-performing CLIP instructor model integration on DFN datasets and improved graphical raw...

InternVL3.5 (Shusheng-Wanxiang 3.5) is an open source multimodal large model of the Shanghai Artificial Intelligence Laboratory, the model is fully upgraded in terms of general ability, reasoning ability and deployment efficiency, providing nine sizes of versions from 1 billion to 241 billion parameters, covering different resource demand scenarios, including thick...

FastVLM (Fast Vision Language Model) is an efficient visual language model introduced by Apple Inc. With FastViTHD hybrid visual coder as the core, it incorporates convolutional and Transformer architectures to significantly reduce visual...

Meeseeks is an open source large model evaluation set used by the Meituan M17 team to evaluate the model's ability to follow instructions.Meeseeks uses a three-tiered evaluation framework to comprehensively measure whether the model is able to generate answers in strict accordance with the user's instructions from the macro to the micro level, without evaluating the knowledge of the content of the answers positively ...