Qwen Image 2.1 Studio by MariucaFuckingGood

Qwen Image 2.1 Studio by MariucaFuckingGood

Puts nine image tools for Qwen Image 2.1 in one page, covering generation, editing, object removal, relighting and outpainting. It suits designers and photo editors who want one place for AI image edits.

Made by MariucaFuckingGood on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Space gathers nine tools for the Qwen Image 2.1 model in a single Gradio app. The set includes image generation and editing, node-based workflows, taking objects out, orbiting the viewpoint, moving objects, swapping faces, relighting, outpainting and swapping outfits. The interface starts in Spanish and can switch to English. It runs on zero-a10g hardware at Hugging Face.

Who it suits: It suits people who edit or create pictures and want several AI tools behind one interface, with a model choice for each tool.

Industry | Who uses it | How they use it | Example
Design | Graphic designers | Generate and edit images in one place | Remove a distracting object from a mockup
Photography | Photo editors | Relight or extend a photo | Widen a picture beyond its borders with outpainting
Fashion and retail | Product and catalog teams | Try outfit changes on pictures | Swap the outfit on a model photo
AI hobby | Model tinkerers | Compare LoRA and turbo setups | Run the same prompt with two model profiles

You can skip it if you need a service with a private pipeline, since uploads are handled inside a public demo Space.

Where to use it: You can use it in these places:

• A browser: open the Space's page on Hugging Face and nothing needs installing.
• Windows, macOS or Linux: run your own copy with Python 3.12 and a suitable GPU.
• A dedicated GPU Space or your own server, where the app adjusts to what it finds.

Use case | Who | What they get back
Image editing | Designers | An edited picture, for example with an object removed
Relighting | Photographers | A picture with changed lighting
Outpainting | Illustrators | A picture extended past its original edges

How to set it up: You need these first for a local copy:

• Python 3.12 and pip
• A GPU with enough memory for the BF16 base model, the official Turbo generator and the adapters
• Disk space and time for the first start, which downloads the pinned base model and LoRA checkpoints

The online page needs none of this. For your own copy, follow the steps below.

1. Install the Hugging Face tools.
2. Download the Space's files into a folder.
3. Move into that folder.
4. Install Gradio at the version the Space uses, then the other packages.
5. Start the app.

pip install -U huggingface_hub
hf download MariucaFuckingGood/Qwen-Image-2.1-studio --repo-type space --local-dir Qwen-Image-2.1-studio
cd Qwen-Image-2.1-studio
pip install gradio==6.28.0
pip install -r requirements.txt
python app.py

How to use it: Open the Space in a browser, or start your own copy with the command below. Then choose a tool, load a picture or write a prompt, pick a model profile and run it.

python app.py

Tool | What it does
Generation and editing | Makes and edits images from prompts
Workflow nodes | Chains steps together, with Hugging Face sign-in for your own quota
Object removal | Takes an object out of a picture
Viewpoint orbit | Changes the viewpoint around a scene
Moving objects | Moves an object to another place in the picture
Face swapping | Swaps a face in a picture
Relighting | Changes the lighting of a picture
Outpainting | Extends the picture beyond its borders
Outfit swapping | Changes the outfit on a person in a picture

Example: You open the object removal tool, load a photo and pick the object you want gone. After you run it, the app returns the photo with that object removed.

For newcomers and experts: If you are new to this, open the Space and try one tool, such as outpainting, on a single picture. Keep the default model profile and compare the result with your original.

If you know the field, the app keeps a separate adapter recipe for every tab, so one tool's LoRA does not leak into another. On ZeroGPU it prepares the BF16 base, the unfused LoRAs and the official Turbo checkpoint once, and shares the frozen text encoder. The cloud list holds 17 model profiles, and FP8, NVFP4 and GGUF are left out on purpose.

Open a tool tab
Load a picture or write a prompt
Pick a model profile
The Space picks its pipeline and adapters
You get the new image

Good to know: Generation uses the visitor's ZeroGPU quota. On a CPU Space the editor opens but generation is switched off. Results and uploads sit in temporary storage that a restart clears, and the source asks you to keep private photos away from a public demo.

• License: apache-2 for the studio code, while model weights keep their own licenses and the Qwen research license limits commercial use
• Cost to run: the visitor's ZeroGPU quota on Hugging Face, or your own GPU for a local copy
• Data it reaches: pictures you upload, handled inside the Space, with model files downloaded from Hugging Face
• Not for: commercial use of outputs before you read the model cards

On Hugging Face, 13 November 2026

• Likes: 3
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/MariucaFuckingGood/Qwen-Image-2.1-studio
• Files: https://huggingface.co/spaces/MariucaFuckingGood/Qwen-Image-2.1-studio/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.