GameGen-O is a game video generation model launched by Tencent, based on the Transformer architecture, specifically designed for generating open-world video games. The model can simulate various functions of a game engine, including generating game characters, animations, etc.
mPLUG-DocOwl 2 is a multimodal large-scale language model for multi-page document understanding, developed by the mPLUG team at Alibaba Tongyi Lab. It achieves high-resolution document understanding without relying on Optical Character Recognition (OCR) technology...
SleepFM is an open-source multimodal sleep analysis model from Stanford University. Based on 100,000 hours of sleep data from over 14,000 participants, it provides a comprehensive sleep health assessment by integrating brain activity, electrocardiogram, and respiratory signals...
QA-MDT (Quality-aware Masked Diffusion Transformer) is an open-source music generation model jointly developed by the University of Science and Technology of China and iFlytek. The model generates high-quality, musically rich music based on text descriptions...
Vchitect 2.0, developed by the Shanghai Artificial Intelligence Laboratory, is an upgraded open-source video generation model designed to generate video content that aligns with Chinese culture and Eastern aesthetics. The model supports videos up to 20 seconds long...
AI YouTube Shorts Generator is an open-source AI video editing tool that automatically analyzes long videos using technologies such as GPT-4, FFmpeg, and OpenCV, extracting the most exciting segments and trimming them into short video formats...
PhotoPrism is an open-source AI photo management tool written in Go, designed to provide users with a decentralized photo storage and organization solution. It allows users to run it on their own hardware, giving them complete control...
Claude Dev is an AI programming assistant integrated into Visual Studio Code. Based on Anthropic's Claude 3.5 Sonnet model, it provides developers with automated programming support. Claude Dev automatically processes...
FaceSwap is an open-source AI software used to create deepfake videos and images. Based on deep learning technology, it replaces one person's face with another's in a video or image. FaceSwap supports...
Seed-Music is an AI music generation model launched by ByteDance that transforms 10-second audio recordings by users into complete musical works. It utilizes an autoregressive language model and diffusion methods, based on the user's multimodal input (such as style...).
ConFiner is an innovative video generation framework developed by multiple universities and research institutions. It combines several readily available diffusion model experts to generate high-quality and coherent video content without additional training.
M2UGen is an advanced multimodal music understanding and generation framework jointly developed by Tencent PCG ARC Labs and the National University of Singapore. It combines the capabilities of large language models (LLMs) to process text, images, videos, and more...
Nemotron-Mini-4B-Instruct is an open-source, small language model from NVIDIA, optimized for role-playing, retrieval augmentation generation (RAG), and function call tasks. The model is optimized through distillation, pruning, and quantization techniques...
PoseTalk is an open-source project based on text and audio-based gesture control and motion refinement methods for generating talking head videos in a single pass. It synthesizes talking face videos from images, driving audio, and driving gestures, providing users with...
Follow-Your-Canvas is a high-resolution video extension technology developed by Tencent's Hunyuan team in collaboration with Tsinghua University and other institutions. Through two core designs—distributed processing and layout alignment—it extends video content to any desired size...
Filmfotos, developed by DynamicWang, is a LoRA model based on the Flux architecture, focusing on simulating and reproducing the texture and effects of film. Characterized by a low-saturation, Japanese film-like style, it brings a soft, delicate, and retro feel to photos...
Faster Whisper is a high-performance speech recognition tool based on the OpenAI Whisper model, utilizing the CTranslate2 engine for fast inference. It improves speech transcription speed and reduces memory usage while maintaining high accuracy...
SwiftBrush V2 is a single-step text-to-image diffusion model that achieves performance comparable to multi-step Stable Diffusion models through improved training methods and model fusion techniques. The model utilizes better weight initialization...
Jina-embeddings-v3 is an advanced text embedding model from Jina AI, designed for multilingual data processing and long text context retrieval tasks. The model boasts 570 million parameters and supports up to 8192 topics...
Click2Mask is an advanced image editing technology that allows users to perform localized editing by simply clicking on images, without the need for complex masking or detailed descriptions. It achieves this by dynamically generating masks and combining Mixed Latent Diffusion (BLD)...
DrawingSpinUp is an advanced 3D animation generation technology developed by a research team at City University of Hong Kong. It transforms 2D character drawings into dynamic 3D animations while preserving the style of the original artwork...
Qwen2.5-Coder is a full-series code generation model launched by Alibaba's Qwen team, driving the development of open-source code language models. It performs exceptionally well in tasks such as code generation, code reasoning, and code fixing. The series covers different...
Qwen2.5-Math is an open-source AI mathematical model developed by Alibaba's Qwen team. It's an upgraded version of Qwen2-Math and supports both Chinese and English. The model is pre-trained on large-scale mathematical data and combines CoT, PoT, and TIR inference methods...
Qwen2.5 is the latest and most powerful AI model open-sourced by Alibaba's Tongyi Qianwen team. It features models with various parameter scales, including 0.5B, 1.5B, 3B, 7B, 14B, 32B, and 72B. The model utilizes state-of-the-art large-scale pre-training techniques...