Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > ComfyUI-GGUF-Loader > Qwen-Image Model Loader ⚡

ComfyUI Node: Qwen-Image Model Loader ⚡

Class Name

QwenImageModelLoader

Category
🤖 CCTech/Qwen-Image
Author
chriscoletech (Account age: 0days)
Extension
ComfyUI-GGUF-Loader
Latest Updated
2026-11-15
Github Stars
0.03K

How to Install ComfyUI-GGUF-Loader

Install this extension via the ComfyUI Manager by searching for ComfyUI-GGUF-Loader
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-GGUF-Loader in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

Qwen-Image Model Loader ⚡ Description

Loads Qwen-Image diffusion model, text encoder, and VAE from safetensors or quantized GGUF.

Qwen-Image Model Loader ⚡:

QwenImageModelLoader prepares the three components needed for a Qwen-Image workflow in ComfyUI. It loads the diffusion model, text encoder, and VAE as separate, native ComfyUI objects, so you can connect each component to the appropriate part of your workflow. It supports both safetensors and GGUF files for the diffusion model and text encoder, selecting the appropriate loading method from each file’s extension. GGUF models remain quantized when loaded. This gives you one place to assemble the components while keeping the flexibility to choose their individual files.

Qwen-Image Model Loader ⚡ Input Parameters:

unet_name

Select the Qwen-Image diffusion model file. This is the component that performs the image-generation process, so your selection determines which diffusion model the node loads. The menu combines available files from ComfyUI’s unet and unet_gguf filename lists; the input tooltip identifies models/diffusion_models as the model location. Files ending in .gguf use the GGUF loader, while other selected files use ComfyUI’s standard diffusion-model loader. Options are the filenames currently available in those lists. No fixed options, minimum, maximum, or default value are defined in the source.

clip_name

Select the Qwen-Image text encoder file. The text encoder prepares text for the image workflow, so choosing the intended encoder matters when you connect this node to prompt-processing nodes. The menu combines available files from the clip and clip_gguf filename lists; the input tooltip identifies models/text_encoders as the location. Files ending in .gguf use a GGUF loader configured for qwen_image; other files use ComfyUI’s standard CLIP loader with its Qwen-Image type and configured embeddings directories. Options are the filenames currently available in those lists. No fixed options, minimum, maximum, or default value are defined in the source.

vae_name

Select the Qwen-Image VAE file from the filenames available in ComfyUI’s vae list; the input tooltip identifies models/vae as its location. The VAE is the component you use to convert between the workflow’s image representation and viewable images, so the chosen file affects that stage of your workflow. The node passes your selection to its VAE-loading helper. Options are the filenames currently available in the vae list. No fixed options, minimum, maximum, or default value are defined in the source.

Qwen-Image Model Loader ⚡ Output Parameters:

model

A MODEL object containing the loaded diffusion model. Connect it to downstream nodes that accept a ComfyUI model for image generation. It is a usable model object, not a generated image or a filename. When you select a .gguf diffusion model, the node loads it through its GGUF path and keeps the quantized model quantized.

clip

A CLIP object containing the loaded Qwen-Image text encoder. Connect it to compatible text-encoding nodes to prepare prompts for the rest of your workflow. The value represents the loaded encoder, not the prompt text or a completed image.

vae

A VAE object containing the loaded VAE. Connect it to compatible VAE encode or decode nodes where your workflow needs to convert image representations. The value is the loaded component, not an image produced by decoding.

Qwen-Image Model Loader ⚡ Usage Tips:

  • Choose diffusion-model and text-encoder files intended for Qwen-Image, along with the appropriate Qwen-Image VAE. The node loads the files you select; it does not verify that the three files belong together.
  • Use the model output for your model-consuming generation nodes, the clip output for compatible prompt-encoding nodes, and the vae output for VAE operations. This node only loads components; it does not generate an image by itself.
  • If you have a GGUF version of the diffusion model or text encoder, select its .gguf filename directly. The node detects that extension for each of those two components independently, so you can choose the available file format for each.
  • If a file does not appear in a menu, check that ComfyUI can find it in the relevant model category: unet or unet_gguf for the diffusion model, clip or clip_gguf for the text encoder, and vae for the VAE. Refresh ComfyUI’s model lists after adding files if needed.

Qwen-Image Model Loader ⚡ Common Errors and Solutions:

No model files appear in a selection menu

  • Explanation: The menu is built from filenames ComfyUI currently finds in the corresponding model category. An empty menu means the required file is not available in that category’s list.
  • Solution: Place the intended file where your ComfyUI installation looks for that component, check any configured model paths, and refresh or restart ComfyUI so it discovers the file.

File not found for the selected unet_name or clip_name

  • Explanation: For a non-GGUF selection, the node resolves the chosen filename with get_full_path_or_raise. Loading can fail if the file was moved, renamed, or removed after it appeared in the menu.
  • Solution: Confirm the file still exists in the correct model category, refresh the menu, and select its current filename.

GGUF diffusion model or text encoder fails to load

  • Explanation: A filename ending in .gguf is sent to the corresponding GGUF loader. The supplied source does not specify that loader’s exact error messages, so the displayed message may vary with the file and installation.
  • Solution: Check that the selected file is a valid GGUF model for the intended component, that its download completed, and that the GGUF loading code is available in your installation. Read the detailed ComfyUI error log to identify the failing file.

Safetensors diffusion model or text encoder fails to load

  • Explanation: A selected file that does not end in .gguf is sent to ComfyUI’s standard diffusion-model or CLIP loader. A damaged file or one that is not suitable for the selected component can fail there; the exact message comes from that loader rather than this node.
  • Solution: Check that you selected a diffusion model for unet_name and a Qwen-Image text encoder for clip_name. Verify the download and try a known-good file in the same category.

VAE loading fails

  • Explanation: The node passes vae_name to its load_vae helper. That helper’s implementation and exact error messages are not included in the supplied source.
  • Solution: Confirm the selected VAE file is present and intact, then try a VAE intended for your Qwen-Image workflow. Use the detailed ComfyUI error log to see what the helper reported.

Downstream nodes report incompatible components

  • Explanation: This loader returns three separate objects, but it does not check whether your selected files match one another or whether a downstream node accepts them.
  • Solution: Confirm that the files are intended for the same Qwen-Image workflow and connect each output to a node that accepts its type: MODEL, CLIP, or VAE.

Qwen-Image Model Loader ⚡ Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-GGUF-Loader
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

Qwen-Image Model Loader ⚡