ClearerVoice-Studio is an open-source speech processing framework from Alibaba DAMO Academy's Tongyi Lab, integrating functions such as speech enhancement, separation, and audio/video speaker extraction. The framework is based on complex domain deep learning algorithms, effectively eliminating...
PaliGemma 2 is a new generation of visual language models (VLMs) launched by Google DeepMind based on the Gemma 2 language model family, serving as an upgrade to the PaliGemma model. It combines the SigLIP-So400m visual encoder with different...
Optimus-1 is an intelligent agent framework developed by Harbin Institute of Technology (Shenzhen) and Pengcheng Laboratory, designed to address the challenges of completing long-term tasks in open-world environments. The framework combines structured knowledge and multimodal experience, enabling intelligent agents to...
Fox-1 is a series of small language models (SLMs) released by TensorOpera, including Fox-1-1.6B and Fox-1-1.6B-Instruct-v0.1. The Fox-1 models are pre-trained on 3 trillion web-fetched documents and have been tested on 5 billion...
Llama 3.3 is a 70-parameter model from Meta AI, a large-scale multilingual pre-trained language model with performance comparable to the 40-parameter Llama 3.1. The model is specifically optimized for multilingual dialogue and supports English, German, French, Italian, etc.
Aurora is a new image generation model added by xAI to its AI assistant Grok. Aurora excels at creating realistic images, particularly portraits. It can generate images of public and copyrighted figures (such as Mickey Mouse). Aurora's...
One Shot, One Talk is an advanced image generation technology that can generate fully animated, talking avatars with personalized details from a single image, supporting realistic animation effects, including natural facial expressions and vivid body movements...
GLM-4V-Flash is the first free multimodal model API launched by Zhipu AI. The GLM-4V-Flash model possesses advanced image processing capabilities such as image description generation, image classification, visual reasoning, visual question answering (VQA), and image sentiment analysis...
Clone-voice is an open-source voice cloning tool that uses deep learning technology to analyze and simulate human voices, achieving high-quality voice cloning. The tool supports 16 languages, including Chinese, English, Japanese, and Korean, and can...
SNOOPI is an innovative text-to-image generation framework that enhances model performance and control through guidance based on an augmented single-step diffusion model. SNOOPI includes PG-SB (appropriate guidance - SwiftBrush) and NASA (negative away-turning attention...
MEMO (Memory-Guided EMOtionaware diffusion) is an audio-driven portrait animation framework developed by Skywork AI, Nanyang Technological University, and the National University of Singapore. It is used to generate images with consistent identity and expressiveness...
NVILA is a series of visual language models from NVIDIA that balances efficiency and accuracy. The model uses a "expand first, compress later" strategy to effectively handle high-resolution images and long videos. NVILA undergoes system optimization during training and fine-tuning...
O1-CODER is an open-source project launched by a research team at Beijing Jiaotong University. It aims to replicate OpenAI's O1 model, focusing on coding tasks. O1-CODER combines reinforcement learning (RL) and Monte Carlo tree search (MCTS) techniques to improve...
SOLAMI is an innovative VR-based 3D role-playing AI system developed by a research team at Nanyang Technological University. It allows users to engage in immersive interactions with virtual characters using voice and body language, based on a social vision-language-behavior model...
Florence-VL is an innovative multimodal large-scale language model (MLLM) jointly developed by the University of Maryland and Microsoft Research. Florence-VL enriches visual representations with the generative visual foundation model Florence-2, enabling it to capture image...
OOTDiffusion is an open-source AI virtual try-on tool that intelligently adapts to different genders and body types, automatically adjusting clothing size and shape to generate a natural-fitting look. OOTDiffusion supports half-body and full-body try-on modes...
The Willow quantum chip, Google's latest offering with 105 physical qubits, represents a significant breakthrough in quantum error correction and computational efficiency. Willow successfully solves a nearly 30-year-old error correction problem in quantum computing, achieving…
GenMAC is an iterative framework for multi-agent collaboration developed by the University of Hong Kong, Tsinghua University, and Microsoft Research, used to solve complex scene generation problems in text-to-video conversion. It is based on breaking down the task into design, generation, and redesign...
Amurex is an AI-powered meeting assistant that helps users improve meeting efficiency through features such as real-time suggestions, intelligent summarization, key point extraction, late join review, and complete meeting transcripts. As an open-source tool, Amurex emphasizes transparency...
EasyHealth is a WeChat mini-program developed by faculty and students of the Health Communication Department at Peking University. It's an intelligent Q&A system for health education built on large language model technology. The program offers AI-powered intelligent Q&A, emotional intelligence algorithms, health records, and more...
See3D (See Video, Get 3D) is a 3D generative model developed by the Beijing Academy of Artificial Intelligence. It learns from large-scale unlabeled internet videos to generate 3D content from them. This differs from traditional camera-based models...
VISION XL is a high-efficiency video restoration and super-resolution tool based on latent diffusion model technology, focusing on solving the inverse problem of high-definition video. The tool can repair missing parts of videos, remove blur, and significantly improve video clarity, up to...
SPDL (Scalable and Performant Data Loading) is an open-source data loading tool launched by Meta AI, which improves the efficiency of AI model training. Based on multi-threading technology, it achieves high-throughput data loading and reduces computational overhead...
CausVid is an autoregressive real-time video generation technology jointly developed by Adobe and MIT, enabling instant video playback. It builds an autoregressive generation model based on a distillation-pretrained bidirectional diffusion model, reducing video generation latency...