Home Creators Posts Import Register Favorites Logout

Content

Qwen Image Edit full tutorial with 26 different demo cases, from photo restoration and text changes to stylization, object removal and outpainting in SwarmUI.

Prompts and demo images are shared, with step-by-step SwarmUI and ComfyUI setup, low-VRAM GGUF models and a fix for black images. The demos show Qwen Image Edit outperforming FLUX Kontext Dev; Massed Compute and RunPod tutorials are linked.

Tutorial Link

https://youtu.be/gLCMhbsICEQ

Extra Info

I tested newly arrived Qwen-Image-Edit-Lightning-8steps-V1.0 (arrived after tutorial recorded) and definitely our existing preset which uses Qwen-Image-Lightning-8steps-V1.1 is better than it, so this tutorial and presets are still 100% best quality and up-to-date

Info

Qwen Image Edit just has been published and since then I have been experimenting to prepare you this amazing tutorial. I have literally shown 26 unique cases and provided demo images and prompts. After watching this tutorial your image editing skills will move to next level i promise you that. Also this tutorial will give you a lot of ideas.

Content

Video Chapters

  • 0:00 Introduction to the Qwen Image Edit Model & Tutorial Overview

  • 0:32 Demo 1: Next-Level AI Photo Restoration

  • 1:04 Demo 2: Changing Text on Existing Images

  • 1:22 Demo 3: Image Stylization (Pixel Art, Cartoon, Pixar, 3D Render)

  • 2:21 Demo 4: Changing Object Materials (Wood to Marble)

  • 2:41 Demo 5: Generating High-Quality Logos from a Blank Canvas

  • 3:39 Demo 6: Generating Images with High-Quality Text

  • 3:57 Demo 7: Adding Realistic Objects to a Scene (Lion)

  • 4:21 Demo 8: Removing Objects from an Image

  • 4:46 Demo 9: Replacing People with Characters (Iron Man)

  • 5:25 Demo 10: Adding Accessories to People (Hat)

  • 5:44 Demo 11: Extracting Clothing from an Image

  • 6:03 Demo 12: Rotating Subjects to Different Angles

  • 6:37 Demo 13: Anime & Ghibli Stylization

  • 7:11 Demo 14: Background Removal & Replacement

  • 7:30 Demo 15: Image Expansion (Outpainting)

  • 7:47 Demo 16 (Limitation): Modifying Facial Expressions

  • 8:16 Demo 18: Changing Clothing (T-Shirt to Suit)

  • 8:34 Demo 19: Changing the Season (Summer to Winter)

  • 9:08 Demo 21: Keeping Text While Changing Background & Style

  • 10:01 Demo 22: Advanced Text & Background Styling for Thumbnails

  • 10:42 Demo 24: Colorizing & Shading Rough Sketches

  • 11:07 Demo 25: Rendering Professional & Technical Sketches

  • 11:37 Demo 26: Converting Real Photos into Professional Sketches

  • 12:04 Installation Guide: Required Downloads & Initial Setup

  • 12:50 Step 1: Updating ComfyUI & SwarmUI

  • 13:28 Step 2: Importing the Latest Presets (Clean Import Method)

  • 14:21 Step 3: Downloading the Qwen Image Edit Models

  • 14:44 How to Select Low VRAM Models (GGUF Quantizations)

  • 15:46 Troubleshooting: How to Fix Black Image Outputs

  • 16:52 How to Use: Applying the Preset & Verifying Model Architecture

  • 18:20 Workflow: Using a Prompt Image (No Init Image)

  • 19:22 CRITICAL STEP: Matching Output Resolution to Input Image

  • 20:00 Tutorial: How to Perfectly Resize Images with Paint.net

  • 21:39 Running Your First Generation & Checking Progress

  • 22:54 How to Upscale Your Generations

  • 23:45 When to Use (and Not Use) the Upscaler

  • 26:03 Generating Images from a Blank Canvas (No Input Image)

  • 28:36 Comparing Upscaled vs. Original Generations

  • 30:00 For Best Realism: Use the SUPIR Upscaler

  • 31:11 Coming Soon: Qwen Image Edit LoRA Training Application

  • 32:14 Final Resources & Where to Get Help

20 Unique Cases Demo Images Comparisons

More Info From Official Source

Introduction

We are excited to introduce Qwen-Image-Edit, the image editing version of Qwen-Image. Built upon our 20B Qwen-Image model, Qwen-Image-Edit successfully extends Qwen-Image’s unique text rendering capabilities to image editing tasks, enabling precise text editing. Furthermore, Qwen-Image-Edit simultaneously feeds the input image into Qwen2.5-VL (for visual semantic control) and the VAE Encoder (for visual appearance control), achieving capabilities in both semantic and appearance editing. To experience the latest model, visit Qwen Chat and select the "Image Editing" feature.

Key Features:

  • Semantic and Appearance Editing: Qwen-Image-Edit supports both low-level visual appearance editing (such as adding, removing, or modifying elements, requiring all other regions of the image to remain completely unchanged) and high-level visual semantic editing (such as IP creation, object rotation, and style transfer, allowing overall pixel changes while maintaining semantic consistency).

  • Precise Text Editing: Qwen-Image-Edit supports bilingual (Chinese and English) text editing, allowing direct addition, deletion, and modification of text in images while preserving the original font, size, and style.

  • Strong Benchmark Performance: Evaluations on multiple public benchmarks demonstrate that Qwen-Image-Edit achieves state-of-the-art (SOTA) performance in image editing tasks, establishing it as a powerful foundation model for image editing.

License Agreement

Qwen-Image is licensed under Apache 2.0.

Related posts

All posts in this topic: AI Image Generation and Editing Apps · New to SECourses? Start Here: Best and Most Popular SECourses Posts

Files

Qwen Image Edit Full Tutorial: 26 Different Demo Cases, Prompts & Images, Pwns FLUX Kontext Dev

Qwen Image Edit just has been published and since then I have been experimenting to prepare you this amazing tutorial. I have literally shown 26 unique cases and provided demo images and prompts. After watching this tutorial your image editing skills will move to next level i promise you that. Also this tutorial will give you a lot of ideas. Prompts, Demo images, Models downloader, SwarmUI installer and Presets : https://www.patreon.com/posts/114517862 ComfyUI backend installer : https://www.patreon.com/posts/105023709 Image and Video comparison sliders : https://www.patreon.com/posts/133935178 How to install ComfyUI and SwarmUI main tutorial : https://youtu.be/fTzlQ0tjxj0 Massed Compute SwarmUI & ComfyUI tutorial : https://youtu.be/8cMIwS9qo4M Massed Compute is recommend cloud service RunPod SwarmUI & ComfyUI tutorial : https://youtu.be/R02kPf9Y3_w SECourses Discord (11k+ members) : https://discord.com/servers/software-engineering-courses-secourses-772774097734074388 SECourses Reddit (6k+ members) : https://www.reddit.com/r/SECourses/ Video Chapters 0:00 Introduction to the Qwen Image Edit Model & Tutorial Overview 0:32 Demo 1: Next-Level AI Photo Restoration 1:04 Demo 2: Changing Text on Existing Images 1:22 Demo 3: Image Stylization (Pixel Art, Cartoon, Pixar, 3D Render) 2:21 Demo 4: Changing Object Materials (Wood to Marble) 2:41 Demo 5: Generating High-Quality Logos from a Blank Canvas 3:39 Demo 6: Generating Images with High-Quality Text 3:57 Demo 7: Adding Realistic Objects to a Scene (Lion) 4:21 Demo 8: Removing Objects from an Image 4:46 Demo 9: Replacing People with Characters (Iron Man) 5:25 Demo 10: Adding Accessories to People (Hat) 5:44 Demo 11: Extracting Clothing from an Image 6:03 Demo 12: Rotating Subjects to Different Angles 6:37 Demo 13: Anime & Ghibli Stylization 7:11 Demo 14: Background Removal & Replacement 7:30 Demo 15: Image Expansion (Outpainting) 7:47 Demo 16 (Limitation): Modifying Facial Expressions 8:16 Demo 18: Changing Clothing (T-Shirt to Suit) 8:34 Demo 19: Changing the Season (Summer to Winter) 9:08 Demo 21: Keeping Text While Changing Background & Style 10:01 Demo 22: Advanced Text & Background Styling for Thumbnails 10:42 Demo 24: Colorizing & Shading Rough Sketches 11:07 Demo 25: Rendering Professional & Technical Sketches 11:37 Demo 26: Converting Real Photos into Professional Sketches 12:04 Installation Guide: Required Downloads & Initial Setup 12:50 Step 1: Updating ComfyUI & SwarmUI 13:28 Step 2: Importing the Latest Presets (Clean Import Method) 14:21 Step 3: Downloading the Qwen Image Edit Models 14:44 How to Select Low VRAM Models (GGUF Quantizations) 15:46 Troubleshooting: How to Fix Black Image Outputs 16:52 How to Use: Applying the Preset & Verifying Model Architecture 18:20 Workflow: Using a Prompt Image (No Init Image) 19:22 CRITICAL STEP: Matching Output Resolution to Input Image 20:00 Tutorial: How to Perfectly Resize Images with Paint.net 21:39 Running Your First Generation & Checking Progress 22:54 How to Upscale Your Generations 23:45 When to Use (and Not Use) the Upscaler 26:03 Generating Images from a Blank Canvas (No Input Image) 28:36 Comparing Upscaled vs. Original Generations 30:00 For Best Realism: Use the SUPIR Upscaler 31:11 Coming Soon: Qwen Image Edit LoRA Training Application 32:14 Final Resources & Where to Get Help Qwen-Image-Edit: Revolutionizing AI-Powered Image Editing Research indicates Qwen-Image-Edit, released by Alibaba's Qwen team, excels in semantic and appearance edits, offering precise control via text prompts. It seems likely that its bilingual text editing capabilities make it ideal for multilingual content creation, though challenges like iterative refinements may arise in complex tasks. Evidence leans toward it outperforming models like FLUX.1 in benchmarks, but open-source nature invites community tweaks for broader applications. Core Features Built on a 20B-parameter foundation, Qwen-Image-Edit uses dual encoding—Qwen2.5-VL for semantics and VAE for visuals—to enable edits like object addition/removal, style transfer, and text modifications while preserving fidelity. It supports chained editing for progressive corrections, such as fixing calligraphy errors. Performance and Use Cases Benchmarks show state-of-the-art results, e.g., 7.56 on GEdit-Bench-EN. Practical uses include IP design (e.g., themed emojis) and professional photography enhancements. Available under Apache 2.0 on Hugging Face and GitHub. Potential Limitations While powerful, it may require hardware for local runs and performs best with refined prompts. Community feedback highlights excitement for integrations like ComfyUI. Some background music by NoCopyrightSounds : https://gist.github.com/FurkanGozukara/681667e5d7051b073f2e795794c46170

Comments

Juanmyth

Fantastic. Looking forward to the training results and tutorial.

BecauseReasons

Since the tutorial mentioned that you can expand/outpaint but now "how"... Does anyone know/figured it out?

Rudbby

I am getting the follwing error aftert following the tutorial: All available backends failed to load the model 'D:/SwarmUI_Model_Downloader_v77/SwarmUI/Models/diffusion_models/Qwen_Image_Edit_GGUF_Q8_0.gguf'. Possible reason: ComfyUI errored: { "error": { "type": "prompt_outputs_failed_validation", "message": "Prompt outputs failed validation", "details": "", "extra_info": {} }, "node_errors": { "100": { "errors": [ { "type": "required_input_missing", "message": "Required input is missing", "details": "type", "extra_info": { "input_name": "type" } } ], "dependent_outputs": [ "9" ], "class_type": "CLIPLoader" } } }

SECourses

it is easy. shown in tutorial. it also exists in the demo zip file and prompt there too

Rudbby

Yeah I was using v77 of swarm downloader and my swarm back end was a new install of comfy v49. Both are fully updated, and if I go to the comfy tab I can generate there, but error every I time in the generate tab.

SECourses

in that case your metadata etc is broken / corrupted. check each model and reset their metadata. try this hit utilities->Reset All Metadata

Mr. Friendly

I saw that qwen now has controlnet, but I seem to fail with getting them to work in swarm ui. Is there something I do or is it just not supported yet? Anyone have an idea?

SECourses

it is on my research list i saw too

cool1

The video was saying to install swam ui nodes by clicking the Windows_Install_Update_SwarmUI_Nodes.bat bat file. I clicked that and then told to restart the server (thinking that might be related to the problem of the Lora bit still not working in the preset for qwen image edit), so after it restarted the server it said: 16:33:04.527 [Warning] [ComfyUI-0/STDERR] Traceback (most recent call last): 16:33:04.528 [Warning] [ComfyUI-0/STDERR] File "H:\ComfyUI-from-SECourses-new-install\ComfyUI\nodes.py", line 2129, in load_custom_node 16:33:04.528 [Warning] [ComfyUI-0/STDERR] module_spec.loader.exec_module(module) 16:33:04.529 [Warning] [ComfyUI-0/STDERR] File "", line 879, in exec_module 16:33:04.530 [Warning] [ComfyUI-0/STDERR] File "", line 1016, in get_code 16:33:04.531 [Warning] [ComfyUI-0/STDERR] File "", line 1073, in get_data 16:33:04.532 [Warning] [ComfyUI-0/STDERR] FileNotFoundError: [Errno 2] No such file or directory: 'H:\\ComfyUI-from-SECourses-new-install\\ComfyUI\\custom_nodes\\SwarmUI_ExtraNodes\\__init__.py' 16:33:04.533 [Warning] [ComfyUI-0/STDERR] why does it seem like clicking the update swarmui nodes bat file has messed up the comfyui installation?

SECourses

well you must report this to swarmui developer. he would fix

cool1

Okay thanks. I assumed you'd done it because in the video you said "I made another update so it will automatically install SwarmUI extra nodes..." and your post said "New Bat file Windows_Install_Update_SwarmUI_Nodes.bat implemented" so I'd assumed you'd implemented it, not someone else. So I assume if anyone has an error with running the .bat files you post they will now need to contact the other developers to try and fix things instead of posting about the error here.

SECourses

yes it installs. but it installs whatever swarmui developer has. if broken it is his thing to fix because i directly install from his repo

cool1

Thanks for the help

Phokage

I followed your tutorial which I think is amazing, there were definitely moments when I had to do some deep diving as it doesn't cover some basic things like which folder does what. I am sure you have it in another video somewhere, I am now generating images which is great. Down side is it 50/50 doesn't use my image that I supplied for it to edit. I am using your prompts unchecked Init image, followed the resolution trick you showed, but for the life of me can't figure out why it won't use my photos. Reaching out to see if I may have missed something. Thanks in advance!

SECourses

it depends on what you are trying to do. at some tasks sadly it is failing. but for my demo prompts it should work. generate at least 4 images with random seed to be sure

Phokage

I've tried different images and also tried your own photos you provided but for whatever reason it just doesn't want to use the photo i provided. It's almost like it's only doing Text to Image instead of Image to Image. (yes I have -1 for random image) :(

SECourses

did you change the model architecture? if you upgrade gold tier i can connect pc and check

Phokage

By that do you mean the preset and model? Model is Qwen_image_edit_GGUF_Q8_0 and preset i chose was the same you mentioned in your video Qwen image edit fast 12 Steps. I do not mind upgrading to the gold tier but I believe if you could point me in the right direction I can be self sufficient and try to solve it myself saving you time. Thanks again Furkan for replying thus far.

SECourses

well if you are following every step of video it should work. sadly i dont know which part you skipped

Craig Empson

How I do install and run SwarmUI on Massed Compute? The ZIP file with the model downloader only seems to have installation instructions for Windows and Runpod.