闲社服务运行正常AI智能体自动化平台
中文
语言 / Language简体中文SelectedEnglish

选择显示语言

Emu35-NF4

wikeeyang/Emu35-NF4

发布机构wikeeyang

下载访问条件魔搭来源

==================================================================================================== 本模型为:https://www.modelscope.cn/models/BAAI/Emu3.5 的 NF4 量化版本,可用官方 inference 代码直接加载,需加装 bitsandbytes 依赖。 模型全部加载到显卡的情况下,需占用 24GB,跑图最大需要 32GB

图片生成
模型详情

模型介绍

中文整理

====================================================================================================

本模型为:https://www.modelscope.cn/models/BAAI/Emu3.5 的 NF4 量化版本,可用官方 inference 代码直接加载,需加装 bitsandbytes 依赖。

模型全部加载到显卡的情况下,需占用 24GB,跑图最大需要 32GB 显存。(根据本人测试情况,安装 flash attn==2.7.4 预编译 whl 也行)

Prompt: "Live shot, close-up, full-body photo, a snow leopard standing on a rock, the body is standing sideways, standing slightly upward on the rock, the tail is slightly cocked, the head is twisted to face the camera, the eyes are looking directly at the camera, the expression is majestic, the background is slightly blurred in the distance, gray rocks and mountains."

Emu3.5-Image 是由国内 BAAI 智源研究院 开源的最新全模态模型,效果比肩 Google-Banana,以下介绍内容引用自官方模型介绍页。本模型为社区研究学习使用,请遵守官方相关版权协议。

====================================================================================================

Emu3.5: Native Multimodal Models are World Learners

Emu3.5 Team, BAAI

Project Page 🤗HF Models Paper

🔹 **Core Concept** **Description**

:-: :--------------------------------------- :-----------------------------------------------------------------------------------------------------------------------------------------

🧠 **Unified World Modeling** Predicts the **next state jointly across vision and language**, enabling coherent **world modeling** and **generation**.

🧩 **End-to-End Pretraining** Trained with a **unified next-token prediction** objective over **interleaved vision–language sequences**.

📚 **Over 10T+ Multimodal Tokens** Pre-trained on **over 10 trillion interleaved tokens** from **video frames** and **transcripts**, capturing **spatiotemporal structure**.

🔄 **Native Multimodal I/O** Processes and generates **interleaved visual–text sequences** without **modality adapters** or **task-specific heads**.

🎯 **RL Post-Training** Large-scale **reinforcement learning** enhances **reasoning**, **compositionality**, and **generation quality**.

⚡ **Discrete Diffusion Adaptation (DiDA)** Converts **sequential decoding → bidirectional parallel prediction**, achieving **≈20× faster inference without performance loss**.

🖼️ **Versatile Generation** Excels in **long-horizon vision–language generation**, **any-to-image (X2I)** synthesis, and **text-rich image creation**.

🌐 **Generalizable World Modeling** Enables **spatiotemporally consistent world exploration**, and **open-world embodied manipulation** across diverse scenarios.

🏆 **Performance Benchmark** Matches **Gemini 2.5 Flash Image (Nano Banana)** on **image generation/editing**, and **outperforms** on **interleaved generation tasks**.

Table of Contents

  • Model & Weights
  • Quick Start
  • Schedule
  • Citation
  • Model & Weights

Model name HF Weight

  • ----------------------- ---------

Emu3.5 🤗 HF link

Emu3.5-Image 🤗 HF link

Emu3.5-VisionTokenizer 🤗 HF link

  • Quick Start

Environment Setup

Configuration

Edit configs/config.py to set:

Paths: model path , vq path

Task template: task type in {t2i, x2i, howto, story, explore, vla} , use image controls usage (set to true when reference images are provided)

Sampling: sampling params (classifier free guidance, temperature, top k/top p, etc.)

Run Inference

Protobuf outputs are written to outputs//proto/ . For better throughput, we recommend ≥2 GPUs.

Visualize Protobuf Outputs

To visualize generated protobuf files:

  • Schedule

[x] Inference Code

[ ] Advanced Image Decoder

[ ] Discrete Diffusion Adaptation(DiDA)

  • Citation

优先采用原作者提供的中文模型卡片;仅有英文说明时由闲社AI翻译整理。重要参数、许可证和商用范围请以原文为准。

你将获得什么

完整仓库方式包含该固定版本中的权重、配置、分词器及说明文件;指定版本方式只下载所选文件或整组分片。多模态模型的投影文件、配置等依赖,请按作者说明配齐。

完整仓库可能包含多个权重格式,因此需要的磁盘空间可能大于单一格式的模型大小。下载文件不等于完成模型部署。

使用前须知

运行环境

运行模型需要的设备、工具与依赖,以原作者说明为准。

授权范围

是否允许商用、修改和分发,请查看模型许可证。

闲社提供资料与镜像下载方法,不代表已运行模型或完成安全审计。下载的代码应先检查,再执行。

返回顶部