zhangxiaole258/z-image-turbo-gguf-confyui overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "tags": [
      "stable-diffusion",
      "comfyui",
      "gguf",
      "text-to-image",
      "z-image-turbo",
      "workflow"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "tags": [
        "stable-diffusion",
        "comfyui",
        "gguf",
        "text-to-image",
        "z-image-turbo",
        "workflow"
      ]
    },
    "hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/67410e4f1b318816c2d21ff8/O_H0XKLp7gSxVWfJYiUqn.gif",
    "summary": "This repository contains the GGUF quantized weights for the **z-image-turbo** model, optimized to run in environments with limited VRAM resources (though still demanding) using ComfyUI. The goal of this upload is to enable the execution of this pipeline by leveraging the efficiency of the GGUF format for both the UNET and the Text Encoder (Qwen). Feel free to download and use just the workflow (json) in the models tab and versions! !ezgif-8e3c09e7876162d0 Example Comparison !ILKCwkG5LkjF2ZrAXXRbJ (1) !EpzgxY40FbLEE3oGUBDIi (2) !gNM6MhM7HPIKAj7YZ2bHz (3)",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\ntags:\n- stable-diffusion\n- comfyui\n- gguf\n- text-to-image\n- z-image-turbo\n- workflow\n---\n\n# z-image-turbo (GGUF Version) for ComfyUI\n\nThis repository contains the GGUF quantized weights for the **z-image-turbo** model, optimized to run in environments with limited VRAM resources (though still demanding) using [ComfyUI](https://github.com/comfyanonymous/ComfyUI).\n\nThe goal of this upload is to enable the execution of this pipeline by leveraging the efficiency of the GGUF format for both the UNET and the Text Encoder (Qwen).\nFeel free to download and use just the workflow (json) in the models tab and versions!\n\n![ezgif-8e3c09e7876162d0](https://cdn-uploads.huggingface.co/production/uploads/67410e4f1b318816c2d21ff8/O_H0XKLp7gSxVWfJYiUqn.gif)\n \n Example Comparison\n![ILKCwkG5LkjF2ZrAXXRbJ (1)](https://cdn-uploads.huggingface.co/production/uploads/67410e4f1b318816c2d21ff8/YOjeIsIsy-iOQn6IA6oQ7.png)\n![EpzgxY40FbLEE3oGUBDIi (2)](https://cdn-uploads.huggingface.co/production/uploads/67410e4f1b318816c2d21ff8/ocCTW4kyr0Y6tiKHYyfRq.png)\n![gNM6MhM7HPIKAj7YZ2bHz (3)](https://cdn-uploads.huggingface.co/production/uploads/67410e4f1b318816c2d21ff8/unhVgOc7cWgN3ZIQW0ted.jpeg)\n\n## 📂 Files and Structure\n\nThe models are organized within the repository folders as follows:\n\n* **UNET:** `models/unet/z_image_turbo-Q8_0.gguf`\n    * *Q8 quantized version of the main diffusion model.*\n* **Text Encoder:** `models/text_encoders/Qwen3-4B-UD-Q5_K_XL.gguf`\n    * *Qwen3 4B LLM quantized in Q5, used for prompt processing.*\n* **VAE:** `models/vae/ae.safetensors`\n    * *Standard Variational Autoencoder for decoding the image.*\n\n* **Feel free to download and use just the workflow!**\n\n## ⚙️ Installation in ComfyUI\n\nTo use these models, you will need custom Nodes that support GGUF loading (such as City96's `ComfyUI-GGUF` or similar).\n\n1.  **Download the files:**\n    * Move the `.gguf` file from the `unet` folder to: `ComfyUI/models/unet/`\n    * Move the `.gguf` file from the `text_encoders` folder to: `ComfyUI/models/clip/` (or `text_encoders` depending on your node loader).\n    * Move the `.safetensors` file from the `vae` folder to: `ComfyUI/models/vae/`\n\n2.  **Recommended Nodes:**\n    * Use **UnetLoaderGGUF** to load `z_image_turbo-Q8_0.gguf`.\n    * Use a GGUF-compatible CLIP/Text Encoder Loader to load `Qwen3-4B`.\n\n## 💻 Hardware Requirements and Performance\n\n⚠️ **Warning: This is a heavy workflow.**\n\nEven with GGUF quantization, the model requires considerable hardware due to the size of the text encoder and the UNET.\n\n* **Minimum GPU:** 12GB VRAM (NVIDIA RTX 3060/4070 or higher).\n* **System RAM:** 32GB Recommended (System may need to offload data to RAM).\n\n### Generation Time\nOn a GPU with **12GB VRAM**, the estimated generation time per image ranges between:\n* **15 to 30 seconds.** making considerably fast!\n\n\nModel Information\nCheck out the original model card Z-Image Turbo for detailed information about the model.\n\n## 🔗 Useful Links\n* [Z-Image-Turbo original post](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo)\n* [ComfyUI-GGUF (Github)](https://github.com/city96/ComfyUI-GGUF)\n* [Z-Image-Turbo-GGUF](https://huggingface.co/jayn7/Z-Image-Turbo-GGUF)\n\n\n---\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "stable-diffusion",
    "comfyui",
    "text-to-image",
    "z-image-turbo",
    "workflow",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "imatrix",
    "conversational"
  ],
  "likes": 1,
  "downloads": 423,
  "gated": false,
  "private": false,
  "last_modified": "2026-03-27T05:43:30.000Z",
  "created_at": "2026-03-27T05:43:30.000Z",
  "pipeline_tag": "text-to-image",
  "library_name": ""
}

{
  "_id": "69c61902ccf891b9af218708",
  "id": "zhangxiaole258/z-image-turbo-GGUF-confyui",
  "modelId": "zhangxiaole258/z-image-turbo-GGUF-confyui",
  "sha": "f4b60de8b270042079f44fa63e6423d5ae51b703",
  "createdAt": "2026-03-27T05:43:30.000Z",
  "lastModified": "2026-03-27T05:43:30.000Z",
  "author": "zhangxiaole258",
  "downloads": 423,
  "likes": 1,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-to-image",
  "library_name": "",
  "siblings_count": 6
}