model_data_json/BAAI_Emu3-VisionTokenizer.json

{
    "model_id": "BAAI/Emu3-VisionTokenizer",
    "downloads": 26276,
    "tags": [
        "transformers",
        "safetensors",
        "Emu3VisionVQ",
        "feature-extraction",
        "custom_code",
        "arxiv:2409.18869",
        "license:apache-2.0",
        "region:us"
    ],
    "description": "--- license: apache-2.0 library_name: transformers --- <div align='center'> <h1>Emu3: Next-Token Prediction is All You Need</h1h1> <h3></h3> Emu3 Team, BAAI | Project Page | Paper | 🤗HF Models | github | Demo | </div> <div align='center'> <img src=\" class=\"interpolation-image\" alt=\"arch.\" height=\"80%\" width=\"70%\" /> </div> We introduce **Emu3**, a new suite of state-of-the-art multimodal models trained solely with **<i>next-token prediction</i>**! By tokenizing images, text, and videos into a discrete space, we train a single transformer from scratch on a mixture of multimodal sequences. ### Emu3 excels in both generation and perception **Emu3** outperforms several well-established task-specific models in both generation and perception tasks, surpassing flagship open models such as SDXL, LLaVA-1.6 and OpenSora-1.2, while eliminating the need for diffusion or compositional architectures. <div align='center'> <img src=\" class=\"interpolation-image\" alt=\"comparison.\" height=\"80%\" width=\"80%\" /> </div> ### Highlights - **Emu3** is capable of generating high-quality images following the text input, by simply predicting the next vision token. The model naturally supports flexible resolutions and styles. - **Emu3** shows strong vision-language understanding capabilities to see the physical world and provides coherent text responses. Notably, this capability is achieved without depending on a CLIP and a pretrained LLM. - **Emu3** simply generates a video causally by predicting the next token in a video sequence, unlike the video diffusion model as in Sora. With a video in context, Emu3 can also naturally extend the video and predict what will happen next. ### Quickstart for Autoencoding",
    "model_explanation_gemini": "\"Emu3 is a multimodal model trained with next-token prediction to generate images, videos, and text while excelling in vision-language understanding, outperforming models like SDXL, LLaVA-1.6, and OpenSora-1.2 without relying on diffusion or compositional architectures.\"\n\n### Features:  \n1. **Multimodal generation** (images, videos, text) via next-token prediction.  \n2. **Flexible resolutions/styles** for image generation.",
    "release_year": "2024",
    "parameter_count": null,
    "is_fine_tuned": false,
    "category": "Embedding",
    "api_enhanced": true
}