pdf-craft is a tool for converting PDF files to other formats (such as Markdown and EPUB), focusing on processing scanned book PDF files. pdf-craft supports extracting the main text, filtering out headers, footers, etc.
TaoAvatar is a high-fidelity, lightweight 3D full-body dialogue virtual human technology developed by Alibaba Group's research team. Based on 3D Gaussian sputtering technology, it can generate photorealistic 3D full-body virtual avatars, supporting high-resolution rendering and...
Mureka O1 is the world's first large-scale music reasoning model released by Kunlun Tech, and the world's first music model to introduce "Chain of Thought" (CoT) technology. Mureka O1 incorporates thinking and self-criticism into the reasoning process...
Mureka V6 is the foundational model of Kunlun Tech's AI music creation platform, supporting pure music generation and AI music creation in 10 languages. It incorporates self-developed ICL (in-context learning) technology, resulting in a wider soundstage...
Video-T1 is a video generation technology jointly developed by researchers from Tsinghua University and Tencent. It improves the quality and consistency of generated videos based on Test-Time Scaling (TTS). Traditional video generation models, in training...
Fin-R1 is the first large-scale R1-class inference model in the financial field, jointly launched by Shanghai University of Finance and Economics and Caiyue Xingchen. Based on the Qwen2.5-7B-Instruct architecture with 7B parameters, it utilizes high-quality thought chain data from financial inference scenarios for SF...
Oliva is an open-source voice RAG assistant that combines Langchain and Superlinked technologies. Based on a voice-driven RAG (Retrieval Augmentation Generation) architecture, it helps users search for information in the Qdrant vector database in real time. ...
Qwen2.5-Omni is Alibaba's open-source flagship multimodal model in the Qwen series. It boasts 7B parameters and powerful multimodal perception capabilities, handling text, image, audio, and video inputs, and supporting streaming...
Cosmos-Reason1 is a series of multimodal large language models from NVIDIA, capable of generating physically based responses. Cosmos-Reason1 includes two models: Cosmos-Reason1-7B and Cosmos-Reason1-5...
RF-DETR is a real-time object detection model introduced by Roboflow. RF-DETR is the first real-time model to achieve a mean accuracy (mAP) of 60+ on the COCO dataset, outperforming existing object detection models. RF-DETR combines LW...
Bolt3D is a new 3D scene generation technology jointly developed by Google Research, the VGG team at Oxford University, and Google DeepMind. It's a potential diffusion model that can generate scenes directly from a single GPU in less than seven seconds...
Piece it Together (PiT) is an innovative image generation framework developed by Bria AI and other institutions, specifically designed to generate complete conceptual images from partial visual components. Based on domain-specific prior knowledge, it transforms user-provided fragments...
TripoSR is an open-source 3D generative model jointly developed by Stability AI and VAST. It can quickly generate high-quality 3D models from a single 2D image in less than 0.5 seconds. The model is based on the Transformer architecture and employs a large-scale reconstruction model...
ModelEngine is Huawei's open-source, end-to-end AI development toolchain, centered around three core functions: data enablement, model enablement, and application enablement. It addresses the challenges of time-consuming data engineering, model training, and other processes involved in the industrial deployment of AI.
BizGen is an AI infographic generation tool jointly developed by Tsinghua University and Microsoft Research, focusing on article-level visual text rendering. It can transform long articles into professional-grade infographics and slideshows with a single click, solving the problems of traditional methods...
Ideogram 3.0 is an AI image generation model launched by Ideogram. Ideogram 3.0 achieves a leap forward in image generation quality, possessing high realism, excellent text rendering, and powerful language understanding capabilities, supporting the generation of...
QVQ-Max is a visual reasoning model launched by Alitongyi, and is the official upgrade of QVQ-72B-Preview. QVQ-Max can "understand" image and video content, combining information to analyze, reason, and solve problems. QVQ-Max supports...
TxGemma is a general-purpose artificial intelligence model launched by Google for drug discovery, accelerating the drug development process through AI technology. Developed based on Google's Gemma framework, it can understand regular text as well as chemical substances, molecules, and...
TripoSG is a high-fidelity 3D shape synthesis technology developed by the VAST-AI-Research team, based on a large-scale corrected flow (RF) model. It utilizes a large-scale corrected flow transformer architecture and hybrid supervised training...
TripoSF is a next-generation 3D basic model launched by VAST, breaking through the bottlenecks of traditional 3D modeling in terms of detail, complex structure, and scalability. It employs the SparseFlex representation method, combined with a sparse voxel structure, to model only the details in the area near the object's surface...
DistilQwen2.5-R1 is a miniaturized series of deep inference models launched by Alibaba, based on knowledge distillation technology. It includes models with four parameter scales: 3B, 7B, 14B, and 32B. DistilQwen2.5-R1 reduces the size of ultra-large-scale models (...
PaddleSpeech is an open-source speech processing tool from the Baidu PaddlePaddle team, providing comprehensive speech processing capabilities, including speech recognition, speech synthesis, speaker recognition, and speech translation. PaddleSpeech offers a command-line interface, services, and more...
ChatAnyone is a real-time stylized portrait video generation framework developed by Alibaba Tongyi Labs. It generates portrait videos with rich facial expressions and upper body movements from audio input. It employs an efficient hierarchical motion diffusion model and...
Math24o is an open-source Chinese large-scale model evaluation benchmark, similar to SuperCLUE, designed for high school Olympiad-level mathematical reasoning assessment. It is primarily used to evaluate the mathematical reasoning capabilities of large language models. It utilizes the 2024 national standard...