Phi-4 is a small language model with 1.4 billion parameters launched by Microsoft. It excels in complex reasoning in fields such as mathematics, as well as traditional language processing. Phi-4 prioritizes data quality in its training, incorporating a large amount of synthetic data to improve model performance...
Insight-V is a multimodal model jointly developed by researchers from Nanyang Technological University, Tencent, and Tsinghua University. It enhances the capabilities of large-scale multimodal language models in long-chain visual reasoning. It is based on scalable data generation streams...
SVDQuant is a post-training quantization technique developed by an MIT research team. For diffusion models, it quantizes the model's weights and activation values to 4 bits, reducing memory usage and accelerating the inference process. SVDQuant introduces high-precision low-rank branch absorption...
Meta Motivo is an AI model developed by Meta Inc. that enhances the realism of the metaverse experience. Meta Motivo simulates human behavior by controlling the full-body movements of a virtual humanoid AI, thus enhancing user interaction. The model employs a non-linear...
Step-1o Audio is China's first end-to-end voice model with hundreds of billions of parameters, launched by Step-1o Technology. It has powerful emotion perception capabilities, accurately identifying the emotions in a user's tone of voice and providing thoughtful responses based on the context.
360gpt2-o1 is a self-developed AI model by 360, demonstrating significant improvements in reasoning capabilities, particularly excelling in mathematical and logical reasoning tasks. The model is optimized through synthetic data, post-training, and the "slow thinking" paradigm...
POINTS 1.5 is a multimodal large model released by Tencent WeChat, an upgrade from POINTS 1.0. The model continues to use the LLaVA architecture from POINTS 1.0, consisting of a visual encoder, a projector, and a large language...
The K1 Visual Thinking Model is a reinforcement learning AI model in Kimi's K1 series. It natively supports end-to-end image understanding and thought chain technology, extending its capabilities to more fundamental scientific fields beyond mathematics. The K1 model excels in image understanding, mathematics...
FreeScale is an inference framework developed by Nanyang Technological University, Alibaba Group, and Fudan University that requires no fine-tuning and enhances the ability of pre-trained diffusion models to generate high-resolution images and videos. FreeScale is based on processing and fusing different...
Freestyler is a rap music generation model jointly developed by the Audio, Speech and Language Processing Group (ASLP@NPU) of the School of Computer Science at Northwestern Polytechnical University, Microsoft, and the Big Data Institute of the Shenzhen Research Institute of the Chinese University of Hong Kong. It can directly...
SnapGen is a text-to-image (T2I) diffusion model jointly developed by Snap Inc., Hong Kong University of Science and Technology, the University of Melbourne, and other institutions. It can quickly generate high-resolution (1024x1024 pixels) images on mobile devices, and only requires...
CosyVoice is a large-scale natural speech synthesis model open-sourced by Alibaba's Tongyi Lab. The model supports five languages: Chinese, English, Japanese, Cantonese, and Korean. It can replicate similar timbres, including intonation and emotional expression, with just 3 to 10 seconds of audio samples...
Megrez-3B-Omni, launched by Wuwen Chip, is the world's first open-source edge-side full-modal understanding model, capable of processing image, audio, and text data. Megrez-3B-Omni outperforms the 34B model on multiple mainstream test sets...
Veo 2 is an AI video generation model from Google DeepMind that can generate high-quality video content based on text or image prompts. Veo 2 supports video production up to 4K resolution, understands camera control commands, and can simulate...
RDT (Robotics Diffusion Transformer) is the world's largest foundational model for diffusing maneuvers in dual-arm robots, developed by the TSAIL team at Tsinghua University's AI Research Institute. RDT boasts one billion parameters and can perform maneuvers without human intervention...
Apollo is a large-scale multimodal model (LMM) developed in collaboration between Meta and Stanford University, focusing on video understanding. Based on systematic research, Apollo reveals the key driving factors for video understanding in LMMs and introduces the "Scaling Constraints"...
BrushEdit is an advanced image editing framework jointly developed by Tencent and Peking University, among other institutions. It is an advanced iteration of the BrushNet model. The framework combines multimodal large language models (MLLMs) and a two-branch image inpainting model...
Bocha Semantic Reranker is a semantic ranking model developed by Bocha AI that improves the accuracy of search results in search and RAG applications. The Bocha Semantic Reranker model is based on text semantics and performs initial ranking...
Ruyi is a large-scale video model developed by TuSimple Technology Co., Ltd. Designed specifically for running on consumer-grade graphics cards, it supports multi-resolution and multi-duration video generation, and features first-frame and first-and-last-frame control, motion control, and camera movement...
Large Action Models (LAMs) are a development framework launched by Microsoft for large action models. They are intelligent systems capable of executing real-world actions, surpassing traditional LLMs (Large Language Models)...
Leffa (Learning Flow Fields in Attention) is a framework developed by Meta AI for generating controllable human images. It incorporates flow field learning into an attention mechanism to precisely control the appearance and pose of a person. Leffa...
LatentLM is a multimodal generative model jointly developed by Microsoft Research and Tsinghua University. It can uniformly process discrete data (such as text) and continuous data (such as images and audio). The model uses a variational autoencoder (VAE) to process continuous data...
Lyra is a high-performance, multimodal, large-scale language model (MLLM) developed by the Chinese University of Hong Kong, SmartMore, and the Hong Kong University of Science and Technology. It focuses on improving the interactive capabilities of speech, vision, and language modalities. Lyra is based on open-source large-scale models and multimodal...
Manga Image Translator is an open-source comic book image text translation tool that can translate text in comics and images with a single click. Manga Image Translator uses OCR technology to recognize text and combines it with machine translation to convert the text...