ClotheDreamer is a 3D clothing generation technology jointly developed by Shanghai University, Shanghai Jiao Tong University, Fudan University, and Tencent YouTu Lab. It can generate high-fidelity, wearable 3D clothing assets based on text descriptions. ClotheDreamer...
Director is a framework for building intelligent video agents. Users can execute complex video tasks using natural language commands, such as video search, editing, compositing, and generation, and stream the results in real time. Based on VideoDB, it's a "video-as-data" platform...
DeepSeek-V2.5-1210 is the final AI model in the DeepSeek V2 series released by DeepSeek, and the final fine-tuned model of DeepSeek V2.5. The model is based on post-training iterations and has been tested in mathematics, programming, writing, and role-playing...
EXAONE 3.5 is an open-source AI model released by LG AI Research Institute, comprising three versions with 2.4 billion, 7.8 billion, and 32 billion parameters. EXAONE 3.5 excels in long text processing, demonstrating superior performance in benchmark tests, particularly in practical applications and long text processing...
DrivingDojo is a dataset developed by the Institute of Automation, Chinese Academy of Sciences, in collaboration with Meituan's autonomous vehicle team. It's used to train and research complex interactive world models for autonomous driving. The dataset contains 18,000 video clips, covering a wide range of topics...
FineWeb 2 is a multilingual pre-trained dataset from Hugging Face, covering over 1000 languages. FineWeb 2 is based on a customized data pipeline, including language recognition, deduplication, content filtering, and PII anonymization, suitable for...
Gemini 2.0 is Google's latest AI model with native multimodal input/output. Gemini 2.0 Flash is the first model in the 2.0 family, centered around multimodal input/output and agent technology, and is twice as fast as the 1.5 Pro...
Shusheng·Wanxiang InternVL 2.5 is an open-source series of large-scale multimodal language models (MLLMs) developed by the OpenGVLab team at the Shanghai AI Lab. This series of models significantly enhances upon InternVL 2.0, particularly in...
ChatTTSPlus is an extended version of ChatTTS, leveraging advanced technologies such as TensorRT acceleration, speech cloning, and mobile model deployment to enhance the performance and flexibility of speech synthesis. On the Windows platform, it achieves over 3 times the performance...
MMAudio is an advanced video-to-audio synthesis technology based on multimodal joint training, allowing the model to be trained on a wide range of audiovisual and audio-text datasets. The core of the technology is the synchronization module, which ensures that the generated audio and video frames are highly synchronized...
Ultravox is a novel multimodal large language model (LLM) that can directly understand text and human speech without relying on a separate automatic speech recognition (ASR) stage. It utilizes multimodal projector technology to convert audio data into high-resolution audio...
Project Mariner is a browser assistant developed by Google DeepMind. Based on Gemini 2.0 technology, Project Mariner uses a Chrome extension to automate browser tasks, understanding and executing web page actions. Proj...
TEN Agent is an open-source real-time multimodal AI agent framework that integrates OpenAI Realtime API and RTC technology. TEN Agent enables multimodal interaction via voice, text, and images, supports high-performance real-time communication, and features low latency...
FLOAT is an audio-driven speaker avatar generation model developed by DeepBrain AI and the Korea Advanced Institute of Science and Technology (KAIST). Based on a flow matching generation model, it learns a motion latent space to achieve efficient temporally consistent motion design. The model is based on...
SynCamMaster is the world's first multi-view video generation model, developed through a collaboration between researchers from Zhejiang University, Kuaishou Technology, Tsinghua University, and the Chinese University of Hong Kong. It can combine 6-DOF camera poses to generate open-ended videos from any viewpoint...
STIV (Scalable Text and Image Conditioned Video Generation) is a large-scale video generation model launched by Apple. STIV boasts 870 million parameters and can handle text-to-video (T2V) and text-to-image...
DiffSensei is a comic generation framework jointly developed by researchers from Peking University, the Shanghai AI Lab, and Nanyang Technological University. It can generate controllable black-and-white comic panels. DiffSensei integrates a diffusion-based image generator and multiple...
Maya is an open-source, multilingual, multimodal model that uses instruction-based fine-tuning to extend its capabilities across multiple languages and cultural contexts. Maya is built on the LLaVA framework and includes a newly created pre-trained dataset containing eight languages, improving visual...
QianYing is a large-scale audio game generation model launched by Giant Interactive, comprising the game video generation model YingGame and the video dubbing model YingSound. YingGame is designed for open-world games and is Giant Interactive's AI platform...
Promptic is a lightweight LLM application development framework that provides an efficient and Python-style development approach. Based on LiteLLM, Promptic allows developers to easily switch between different LLM service providers with just a single line of code change...
SwiftEdit is a text-guided image editing tool developed by the VinAI Research team. Based on innovative one-step diffusion technology, it can achieve fast and high-quality image editing in 0.23 seconds. The core advantage of this tool lies in its one-step...
ChatMCP is an AI chat client based on the Model Context Protocol (MCP), supporting interaction with various large language models (LLMs) such as OpenAI, Claude, and OLLama. ChatMCP features automated MCP installation...
The Multimodal Live API is a low-latency, two-way interactive AI interface from Google that supports text, audio, and video input and outputs audio and text. It helps developers build systems with real-time audio and video streaming capabilities...
CodeArena is an online platform that allows multiple large language models (LLMs) to simultaneously build the same application, displaying real-time ranking results and comparing the code generation capabilities of different LLMs. The CodeArena platform primarily evaluates and compares different LLMs...