Clapper is a free and open-source visual AI video editing tool developed by Julian Bilcke, an AI front-end engineer at HuggingFace. It integrates various generative AI technologies, allowing users to interact, iteratively, and intuitively...
UniTalker is an audio-driven 3D facial animation generation model that can generate realistic facial movements based on input audio. It employs a unified multi-head architecture model, uses datasets with different annotations, and supports multiple languages and audio formats...
Grok-2 is a next-generation AI model from xAI, offering superior chat, programming, and reasoning capabilities. In academic benchmark tests, Grok-2 outperforms its predecessor, Grok-1, in areas such as GPQA, MMLU, MMLU-Pro, and MATH...
MagicPose is an AI video generation model jointly developed by the University of Southern California and ByteDance. It generates realistic videos of human movements and facial expressions without any fine-tuning. MagicPose employs a novel two-stage training strategy...
Agent Q is a self-supervised agent reasoning and search framework developed by MultiOn in collaboration with Stanford University. Agent Q integrates techniques such as Guided Monte Carlo Tree Search (MCTS), AI self-criticism, and Direct Preference Optimization (DPO)...
UniPortrait, launched by Alibaba, is an AI-powered image personalization editing tool that can transform photos into anime styles, supporting group photos and face-swapping technology. UniPortrait utilizes advanced 'ID embedding' and 'ID routing' technologies to...
LabelU is an open-source multimodal data annotation tool that supports image, video, and audio annotation. It features capabilities such as bounding box annotation, polygon annotation, punctuation, line annotation, classification, and description annotation, and can meet the needs of object detection, image classification, etc.
Buzz is an offline speech-to-text tool built on the OpenAI Whisper model, available for Windows, macOS, and Linux systems. Buzz can convert microphone input or audio/video files into text in real time, supporting multiple...
HoloDreamer is an AI text-driven 3D scene generation framework developed by Peking University in collaboration with Pengcheng Laboratory. Through two core modules—stylized panoramic image generation and enhanced two-stage panoramic image reconstruction—it rapidly generates panoramic images from text descriptions...
HumanVid is a high-quality dataset jointly developed by the Chinese University of Hong Kong and the Shanghai Artificial Intelligence Laboratory, specifically designed for training human image animation. It combines real-world video and synthetic data, using carefully designed rules to filter high-quality data...
ORMBG is an open-source AI image segmentation tool initiated by developer schirrmacher on GitHub. ORMBG focuses on accurately removing backgrounds from images, using advanced image processing techniques to achieve precise separation of foreground and background...
Cradle is a multimodal AI agent framework for General Computer Control (GCC), developed by Kunlun Tech in collaboration with the Beijing Academy of Artificial Intelligence, Nanyang Technological University (Singapore), Peking University, and others.
Linly-Dubbing is an open-source AI-powered video dubbing and translation tool that automatically translates video content into multiple languages and generates subtitles. It utilizes WhisperX and FunASR for accurate speech recognition and is based on Edge...
VITA is the world's first open-source multimodal large language model (MLLM) launched by Tencent YouTu Lab. It can understand and process video, images, text, and audio. Based on the Mixtral 8×7B model, it expands the Chinese vocabulary and performs...
AI Scientist is Sakana AI's first fully automated AI system for scientific discovery. It can independently complete the entire research process, from idea generation, coding, experimental execution to writing scientific papers.
ControlNeXt is a novel AI framework for controllable image and video generation, jointly developed by the Chinese University of Hong Kong and SenseTime. It employs lightweight control modules and innovative cross-normalization technology, significantly reducing computational resources and training time...
FancyVideo is an AI-powered text-based video model developed by 360 in collaboration with Sun Yat-sen University. Based on its innovative Cross-frame Textual Guidance Module (CTGM), it can generate dynamic and rich video content based on text prompts...
MovieDreamer is an AI video generation framework developed by Zhejiang University in collaboration with Alibaba specifically for long-form videos. Combining autoregressive models and diffusion rendering technology, it can generate long videos with complex plots and high visual quality.
Glyph-ByT5-v2 is a multilingual visual text rendering project jointly developed by Microsoft Research Asia, Tsinghua University, Peking University, and the University of Liverpool. Glyph-ByT5-v2 supports accurate visual text rendering in 10 different languages...
Llama-3.1-Minitron is an AI model developed in collaboration between NVIDIA and Meta. It's a smaller 4B parameter model refined from the Llama 3.1 8B model through pruning and knowledge distillation techniques. This optimization reduces model size and complexity...
CrewAI is an open-source platform focused on building efficient collaboration among multiple AI agents. By simplifying the use of AI agent frameworks, it enables enterprises to quickly build and deploy intelligent automation solutions. The core advantages of the CrewAI platform lie in...
HybridRAG is a hybrid retrieval-augmented generation architecture developed by Blackstone in collaboration with NVIDIA. It combines a machine learning architecture with a retrieval-augmented generation model. The model searches for...
DeepSeek-Prover-V1.5 is an open-source mathematical model developed by the DeepSeek team, boasting 7 billion parameters. The model combines reinforcement learning (RLPAF) and Monte Carlo tree search (especially a variant of RMaxTS) in mathematical definition...
SadTalker is an open-source AI digital human project jointly launched by Xi'an Jiaotong University, Tencent AI Lab, and Ant Group. SadTalker focuses on generating realistic speaking characters using 3D motion coefficients from a single facial image and audio recording...