gpt-4o-transcribe is a high-performance speech-to-text model from OpenAI. Based on the latest speech model architecture, it is trained with massive amounts of diverse audio data to accurately capture subtle differences in speech and significantly reduce word error rates (...
GPT-4o mini TTS is a lightweight text-to-speech model from OpenAI. It supports converting text into natural, fluent speech, and allows developers to control the tone, emotion, and style of the speech using commands such as "calm," "...".
Futang Baichuan is the world's first large-scale pediatric model, jointly released by the National Children's Medical Center, Beijing Children's Hospital affiliated with Capital Medical University, Baichuan Intelligence, and Xiaofang Health. It provides a comprehensive view of common and complex childhood diseases...
Agent TARS is an open-source multimodal AI agent tool from ByteDance. Based on visual interpretation of web page content, it seamlessly integrates with browsers, command lines, and file systems to plan and execute complex tasks. Agent TARS provides...
gpt-4o-mini-transcribe is a speech-to-text model released by OpenAI, a simplified version of gpt-4o-transcribe. Based on the GPT-4o-mini architecture, gpt-4o-mini-transcribe uses knowledge distillation technology to extract text from large datasets...
SpatialLM is an open-source spatial understanding multimodal model from GroupCore Technology, endowing robots and intelligent systems with human-like spatial cognitive abilities. By analyzing videos captured by ordinary mobile phones, it can reconstruct detailed 3D scene layouts...
Multi-Agent Orchestrator is a framework for managing and coordinating multiple intelligent agents. It identifies the intent of user input through a classifier, assigns requests to the most suitable agent for processing, and utilizes dialogue storage...
LongCat is a generative AI model independently developed by Meituan to improve the company's internal work efficiency and innovation capabilities through artificial intelligence technology. The model possesses powerful multimodal capabilities, able to process various types of data such as text and images...
Food God is Robam Appliances' first large-scale model in the vertical field of cooking. Based on Robam Appliances' 46 years of accumulated massive cooking data, combined with DeepSeek technology, it can accurately understand the needs of cooking scenarios. Users can communicate via voice, text, and other methods...
Soundwave is an open-source speech understanding model from the Chinese University of Hong Kong, Shenzhen, focusing on intelligent alignment and understanding of speech and text. Through innovative alignment and compression adapter technologies, it effectively solves the problem of speech and text...
Motia is an AI Agent framework designed for software engineers, simplifying the development, testing, and deployment of AI agents. It supports multiple programming languages, such as Python, TypeScript, and Ruby, allowing developers to use their proficient...
InfiniteYou (InfU) is an identity-preserving image generation framework developed by ByteDance's Intelligent Creation Team, based on diffusion transformers (such as FLUX). It uses InfuseNet to inject identity features...
RuoYi AI is a full-stack AI development platform that provides complete front-end, back-end management, and mini-program applications, supporting flexible code modification and distribution. RuoYi AI supports local RAG solutions and integrates with Milvus/Weaviate, etc...
MoshiVis is an open-source multimodal speech model developed by Kyutai. Based on the Moshi real-time dialogue speech model, it adds visual input functionality. It enables natural, real-time voice interaction with images, integrating voice and visual information...
DeepMesh is a 3D mesh generation framework proposed by researchers from Tsinghua University and Nanyang Technological University. It generates high-quality 3D meshes based on reinforcement learning and autoregressive transformers. It optimizes mesh generation through two key innovations: one...
BlockDance is a new method for accelerating diffusion models, developed jointly by Fudan University and ByteDance's intelligent creation team. BlockDance reduces redundancy by identifying and reusing structurally similar spatiotemporal features (STSS) in adjacent time steps...
Reve Image is a brand-new AI image generation model from Reve. It focuses on enhancing aesthetic performance, accurate cue following, and excellent typography, generating high-quality visual works. The model exhibits...
StarVector is an open-source multimodal visual language model, jointly developed by ServiceNow Research, Mila-Quebec AI Institute, and ETS Montreal, focusing on converting images and text into scalable vectors...
LHM (Large Animatable Human Reconstruction Model) is a product developed by Alibaba's Tongyi Labs that reconstructs an animated 3D human model from a single image. Based on a multimodal Transformer architecture, it integrates 3D geometry...
OceanDS is China's first large-scale language model in the marine vertical field, jointly launched by the National Marine Environmental Forecasting Center, Ocean Press, and 360 Digital Security Technology. The model utilizes 360 Brain 13B and Deepseek-R1...
Qwen2.5-VL-32B is an open-source multimodal model from Alibaba, with a parameter size of 32 bytes. Based on the Qwen2.5-VL series, the model is optimized using reinforcement learning, resulting in a more human-like response style and significantly improved accuracy...
AndroidGen is a framework developed by the Zhipu Technology team to enhance the capabilities of Large Language Model (LLM)-based agents, especially in situations where data is scarce. The framework collects human task trajectories and trains language models based on these trajectories...
ReasonGraph is an open-source web platform for visualizing and analyzing the reasoning process of Large Language Models (LLMs). ReasonGraph supports over 50 mainstream models (such as Anthropic, OpenAI, Google, etc.), covering...
Gemini 2.5 Pro is Google's latest AI model, a "thinking model" that performs reasoning before responding, improving performance and accuracy. The model excels in multiple benchmark tests, particularly in reasoning and code generation...