StreamBridge is an edge-side video large language model (Video-LLMs) framework jointly developed by Apple and Fudan University, which helps AI understand live video streams in real time. The framework is based on a memory buffer and a round-based attenuation compression strategy...
Step1X-3D is a high-fidelity, controllable 3D asset generation framework developed by StepFun in collaboration with LightIllusions. Based on a rigorous data processing process, it selects 2 million high-quality assets from over 5 million 3D assets...
OpenVision is a family of fully open, efficient, and flexible advanced visual encoders from the University of California, Santa Cruz (UCSC), focusing on multimodal learning. It offers models of various sizes from 5.9M to 632.1M parameters, suitable for...
FLUX-Text is a new multilingual text editing framework launched by Alibaba, based on a diffusion model and a lightweight glyph embedding module. The framework enhances text editing in complex scenarios by injecting glyph conditional information...
MCA-Ctrl (Multi-party Collaborative Attention Control) is an image customization generation framework developed by the Institute of Computing Technology, Chinese Academy of Sciences and the University of Chinese Academy of Sciences. It achieves high-quality image generation based on text and complex visual conditions...
AgentCPM-GUI is an open-source edge GUI agent jointly developed by Tsinghua University and the Mianbi AI team, optimized for Chinese applications. The model is built on MiniCPM-V (8 billion parameters) and can accept smartphone screenshots as input...
AG-UI (Agent-User Interaction Protocol) is an open-source, lightweight, event-based protocol that supports standardized interactions between AI agents and front-end applications. AG-UI defines 16 standard event types...
DreamFit is a virtual fitting framework developed by ByteDance in collaboration with Tsinghua University Shenzhen International Graduate School and Sun Yat-sen University Shenzhen Campus. It's specifically designed for generating human images centered around lightweight clothing. Based on adaptive attention and L...
Minion Agent is an agent framework that supports browser operations, MCP (Multi-Channel Programming), automated planning, and deep research. Minion Agent supports multiple models and offers flexible configuration options to meet diverse needs.
Being-M0 is the first large-scale general motion generation model for humanoid robots, jointly developed by Peking University, Renmin University of China, and other institutions. Being-M0 is based on MotionLib, the industry's first dataset of millions of motion samples, and uses innovative Motion...
DanceGRPO is the first unified visual generative reinforcement learning framework jointly launched by ByteDance Seed and the University of Hong Kong. It applies reinforcement learning to the field of visual generation, covering two major generative paradigms (diffusion and rectified generation)...
AlphaEvolve is a general-purpose scientific agent launched by Google DeepMind. It designs and optimizes advanced algorithms based on a combination of creativity models (LLMs) and automated evaluators. It utilizes both Gemini Flash and Gemini Pro...
WorldMem is an innovative AI world generation model developed by Nanyang Technological University, Peking University, and the Shanghai AI Lab. The model addresses the key issue of inconsistent world generation models over long time periods by incorporating a memory mechanism...
GPDiT (Generative Pre-trained Autoregressive Diffusion Transformer) is a novel video generation model developed by Peking University, Tsinghua University, StepFun, and the University of Science and Technology of China. The model...
Skywork-VL Reward is an open-source multimodal reward model from Skywork AI, providing reliable reward signals for multimodal understanding and reasoning tasks. The model is based on the Qwen2.5-VL-7B-Instruct architecture, and incorporates a reward head structure...
ChatUI is an open-source intelligent conversational UI component library launched by Alibaba. It helps developers quickly build high-quality chat applications, providing features such as responsive design, internationalization, and theme customization. ChatUI is based on Alibaba...
FaceShot is a novel, training-free portrait animation generation framework developed by Tongji University, Shanghai AI Lab, and Nanjing University of Science and Technology. It utilizes an appearance-guided landmark matching module and a coordinate-based landmark relocalization module to generate various character animations...
MT-Color is a controllable image colorization framework based on a diffusion model, jointly developed by Shanghai Jiao Tong University and Bilibili. It achieves accurate instance-level image colorization based on user-provided instance-aware text and masks. The framework is based on pixel-level masks...
Speech-02 is a next-generation text-to-speech (TTS) model from MiniMax. Based on a regressive Transformer architecture, the model achieves zero-shot speech cloning, generating highly similar target speech from just a few seconds of reference speech...
Stable Audio Open Small is a lightweight text-to-audio generation model developed in collaboration between Stability AI and Arm. Based on the Stable Audio Open model, the number of parameters has been reduced from 1.1 billion to 341 million, and the generation speed...
LBM (Latent Bridge Matching) is a novel image-to-image transformation framework developed by the Jasper Research team. It achieves fast and efficient image transformation by constructing bridge matching in the latent space. LBM requires only a single step of inference...
Sketch2Anim is an automation framework developed by the University of Edinburgh in collaboration with Snap Research and Northeastern University. It can directly convert 2D sketch storyboards into high-quality 3D animations. Based on conditional motion synthesis technology, it uses 3D key poses,...
SWE-1 is the first software engineering AI model series launched by the Windsurf team. It is designed to optimize the entire software development process, covering development stages such as code generation, testing, debugging, and data retrieval, thereby improving development efficiency and user experience.
DICE-Talk is a novel emotional dynamic portrait generation framework jointly developed by Fudan University and Tencent YouTu Lab. It supports the generation of dynamic portrait videos with vivid emotional expression while maintaining identity consistency. DICE-Talk introduces emotional...