PixelDance is ByteDance's latest AI video generation model, employing a DiT architecture and supporting both text-based and image-based video. It can understand complex instructions and generate coherent video clips up to 10 seconds long, covering multi-subject interactions...
Seaweed is an AI video generation model launched by ByteDance, supporting both text-based and image-based video modes. Based on the Transformer architecture and trained using spatiotemporal compression technology, it natively supports multi-resolution output and is suitable for...
PortraitGen is an AI-powered portrait video editing tool developed by a research team at the University of Science and Technology of China. Based on 3D Gaussian sputtering technology and a neural Gaussian texture mechanism, it converts 2D portrait videos into 4D Gaussian fields, achieving high-quality 3D...
MMMLU (Multilingual Large-Scale Multitask Language Understanding) is an open-source dataset launched by OpenAI, designed to evaluate and improve the performance of artificial intelligence models across different linguistic, cognitive, and cultural contexts. MMMLU is built upon...
Llama 3.2 is Meta's latest open-source AI large model series, including small and medium-sized visual language models (11B and 90B parameters) and lightweight plain text models (1B and 3B parameters). Llama 3.2 models are specifically designed for edge computing...
MIMO is a novel AI framework for controllable character video synthesis developed by Alibaba Group's Intelligent Computing Research Institute. Based on spatial decomposition modeling technology, it converts 2D video into 3D spatial code, enabling precise control over characters, actions, and scenes...
Doubao AI video models are two large-scale AI video generation models launched by ByteDance: PixelDance and Seaweed. PixelDance is based on a DiT structure and excels at understanding complex instructions, generating coherent video clips with multi-agent interactions...
Meitu MiracleVision is an AI visual model launched by Meitu Inc., focusing on aesthetic creation, including Eastern aesthetics, portraiture, and commercial design. The model has undergone a comprehensive upgrade in video generation capabilities...
SafeEar is an AI-powered audio spoofing detection framework jointly developed by Zhejiang University and Tsinghua University. It protects user privacy while detecting audio spoofing. It employs a decoupled model based on a neural audio codec to separate the acoustic information of speech from...
PDF2Audio is an open-source tool that converts PDF documents into audio content, making it suitable for creating podcasts, lectures, or summaries. It generates podcast scripts based on OpenAI's GPT model and converts them into audio using text-to-speech technology.
GraphReasoning is an artificial intelligence-based method for transforming large amounts of scientific papers into knowledge graphs. Through structured analysis, it calculates node degree, identifies communities and their connectivity, and assesses the centrality of key nodes to reveal...
Comic Translate is an open-source comic translation tool developed by ogkalu2. It helps users automatically translate comics from around the world, supporting English, Korean, Japanese, French, Simplified Chinese, Traditional Chinese, Russian, and more.
PearAI is an open-source, AI-driven code editor built on Visual Studio Code (VSCode). PearAI integrates AI technology to reduce programming workload and improve development efficiency. PearAI supports developers directly...
EzAudio is a text-to-audio (T2A) generation model jointly developed by Johns Hopkins University and Tencent AI Labs. Based on an efficient diffusion transformer technique, it is used to generate high-resolution audio from text prompts...
FLUX-Controlnet-Inpainting is an image restoration tool launched by Alibaba's Alimama, integrating ControlNet and FLUX.1-dev technologies. The tool inpaints images based on a user-specified mask area...
Rope is an open-source AI face-swapping tool built on InsightFace's inswappper_128 model, providing a user-friendly graphical interface. Users can upload images or videos and complete face-swapping operations in seconds, with excellent results...
Westlake-Omni is the world's first open-source, end-to-end Chinese emotion-based speech interaction model, developed by Westlake Omni. The model employs discrete representation, unifies the processing of text and speech modalities, and places particular emphasis on real-time performance for rapid user response...
AutoGen Studio is an open-source graphical interface tool developed by Microsoft Research, designed to simplify the building, debugging, and evaluation of multi-agent systems. AutoGen Studio provides a web-based interactive interface and Python...
IDIFY is a free and open-source online ID photo generator that uses AI technology to automatically cut out images, helping users quickly generate standard-compliant ID photos. Users simply upload their photos to their browser, select the size and background color, and...
Emu3 is a native multimodal world model developed by the Beijing Academy of Artificial Intelligence (BAAI). It employs BAAI's self-developed multimodal autoregressive technology, jointly trained on images, videos, and text, enabling the model to possess native multimodal capabilities...
CCI 3.0 is a large-scale Chinese internet corpus released by the Beijing Academy of Artificial Intelligence (BAAI), containing a 1000GB dataset and a 498GB high-quality subset, CCI 3.0-HQ. This version expands the data scale by nearly one... compared to CCI 2.0.
MemFree is an open-source hybrid AI search engine that integrates multiple AI models and search engines to provide an efficient and diverse search experience. Users can search and ask questions using text, images, documents, and web pages,...
GarDiff is an innovative virtual try-on technology that uses CLIP and VAE encoding to extract the appearance prior of clothing, combined with a clothing focus adapter and high-frequency detail enhancement algorithms, to generate high-fidelity and richly detailed try-on images.
NeMo is an end-to-end cloud-native framework provided by NVIDIA for building, customizing, and deploying generative AI models. It supports applications such as large language models (LLMs), multimodal models, speech recognition, and text-to-speech (TTS)...