WiseDiag, launched by Hangzhou Zhizhen Technology, is a globally leading medical big data model focused on improving the efficiency and quality of healthcare services through artificial intelligence. The model boasts 73 billion parameters and a 32k context length, based on over...
CustomVideoX is an innovative personalized video generation framework jointly proposed by the University of Science and Technology of China (USTC) and Zhejiang University, among others. It generates high-quality, customized videos using reference images and text descriptions. The framework is based on a video diffusion transformer...
KTransformers is an open-source project launched by the KVCache.AI team at Tsinghua University in collaboration with Qujing Technology. It optimizes the inference performance of large language models and lowers the hardware requirements. KTransformers is based on a GPU/CPU heterogeneous computing strategy, using...
DragAnything is a controllable video generation method based on entity representation, developed by Kuaishou in collaboration with Zhejiang University and the Show Lab of the National University of Singapore. It achieves precise motion control of any object in a video based on simple trajectory input...
GENERator is a generative genome model developed by the AI for Science team at Alibaba Cloud's Apsara Lab, focusing on the design and generation of DNA sequences. The model is based on a Transformer decoder architecture and features 98k...
FoloUp is an open-source AI-powered voice interview platform that helps companies conduct recruitment interviews efficiently. FoloUp can automatically generate customized interview questions based on job descriptions and engage in natural, conversational voice interactions with candidates using AI technology...
ConsistentDreamer is a novel image-to-3D asset generation technology proposed by Huawei's Munich Research Center. It generates view-consistent 3D meshes from a single image. The method utilizes Gaussian optimization guided by multi-view prior images...
Cline is an AI-powered programming assistant integrated into VSCode that improves development efficiency through intelligent methods. It boasts powerful code generation and editing capabilities, quickly creating or modifying code files based on user needs, and providing real-time syntax checking...
T1 (Thinker) is the latest deep thinking model launched by Tencent Hunyuan and has been officially integrated with Tencent Yuanbao. The model focuses on logical reasoning and deep thinking, supports online search functionality, and can retrieve information from internet sources, WeChat official accounts, ...
ImageRAG is an image generation technique based on Retrieval-Augmented Generation (RAG). It improves the generation of rare or unseen concepts in text-to-image (T2I) models by dynamically retrieving relevant images...
X-R1 is a low-cost training framework based on reinforcement learning that accelerates the scaling post-training development of large-scale language models. X-R1 trains 0.5B (500 million parameters) R1-Zero models at extremely low cost...
Step-Video-T2V is an open-source text-to-video pre-trained model developed by the Step-Stars team. It boasts 30 billion parameters and can generate high-quality videos up to 204 frames long. The model is based on a deep-compressed variational autoencoder (Video-...
Unsloth is an open-source tool for fine-tuning Large Language Models (LLMs). By optimizing computational steps and GPU kernels, it significantly improves model training speed and reduces memory usage. Unsloth supports several mainstream LLMs, such as Llama-3 and Mist...
Matrix3D is a novel unified photogrammetry model developed through a collaboration between Nanjing University, Apple Inc., and Hong Kong University of Science and Technology. It can perform multiple photogrammetry sub-tasks within a single model, including pose estimation, depth prediction, and new view merging...
Light-A-Video is a training-free video relighting method jointly developed by Shanghai Jiao Tong University, University of Science and Technology of China, Chinese University of Hong Kong, Hong Kong University of Science and Technology, Stanford University, and Shanghai AI Lab. It is based on a progressive...
Finedefics, developed by Professor Peng Yuxin's team at Peking University, is a fine-grained multimodal large-scale model that improves the performance of multimodal large language models (MLLMs) in fine-grained visual recognition (FGVR) tasks. The model enhances the performance of these models by incorporating object-specific...
Step-Audio is the first product-level open-source voice interaction model launched by the Step-Leap Star team. It can generate expressions with emotions, dialects, languages, singing voices, and personalized styles according to different scenario needs, enabling natural and high-quality interaction with users...
SkyReels-V1 is Kunlun Wanwei's first open-source video generation model for AI-powered short drama creation. Based on fine-tuning of tens of millions of high-quality film and television data sets, it achieves film-level generation of micro-expressions and body movements, supporting 33 levels of detail...
LangFlow is a low-code, visual AI application building tool for quickly building and experimenting with LangChain pipelines. Through a drag-and-drop interface, users can easily create complex AI workflows without writing large amounts of code...
SkyReels-A1 is Kunlun Wanwei's open-source, China's first state-of-the-art (SOTA) facial expression and motion control algorithm based on a video pedestal model. SkyReels-A1 enables more precise and controllable generation of human videos, and...
Mistral Saba is a region-specific AI model developed by the French company Mistral AI, focusing on languages and cultures in the Middle East and South Asia. Despite its small size (24 billion parameters), the model excels at processing Arabic and languages of Indian origin...
Grok 3 is the latest generation of artificial intelligence model launched by xAI, a company founded by Elon Musk. Musk has called it "the smartest AI on Earth," and its reasoning capabilities have surpassed those of ChatGPT and DeepWorks in multiple benchmark tests...
Kimi Latest is a real-time updated AI model launched by Kimi, providing users with a model experience that is synchronized with the Kimi intelligent assistant. It supports a 128k context length and can automatically select 8k, 32k, or 128k models based on input...
Omnitool is an open-source AI desktop environment that provides users with a unified interface for quickly accessing and using various AI models. Omnitool supports local operation and can be installed on Mac, Windows, or Linux devices. Data storage...