Picture published by the creator on Hugging Face.
Qwen-Image-2.1-Turbo: Generate and edit images with an eight-step Qwen model
Added · Updated
An accelerated 7-billion-parameter image model for making pictures from text and editing an existing image.
Qwen says this Turbo checkpoint uses eight denoising steps, ships with its recommended sampling schedule, and works through the QwenImage21Pipeline in Diffusers. The model card includes examples for posters, typography, transparent images, single-image changes and multi-reference composition.
Model details
Original model Qwen/Qwen-Image-2.1-Turbo
Input Text or reference image Output Image
Ways to use it Run locally
An eight-step Turbo checkpoint for text-to-image generation and image editing.
Other source-backed versions are listed here with their role.
Viggle/Qwen-Image-2.1-viggle-turbo · Fewer-step adapter — needs the original modelA distilled adapter loaded on Qwen Image for text-to-image and reference-image editing.unsloth/Qwen-Image-2.1-GGUF · Compressed model files — follow the supported setupA GGUF distribution of Qwen Image with source-specific setup instructions.Qwen/Qwen-Image-2.1 · Original modelGenerate or edit images, including transparent artwork.Comfy-Org/Qwen-Image-2.1 · Packaged for ComfyUI workflowsA ComfyUI distribution of Qwen Image with source-linked workflows.What it does
This is the faster checkpoint of Qwen-Image-2.1: you give it a prompt and it generates an image, or you provide a reference image and ask for an edit. The same pipeline covers text-to-image, local changes, transparent output, typography and compositions that use several references. Its eight-step sampling schedule is bundled into the checkpoint, so the documented Diffusers path does not require you to tune a scheduler by hand; prefix key-value caching reuses the text and reference-image context across those steps. The card publishes output presets from square images through wide 16:9 scenes and tall 9:16 images.
You could use it to…
- Make a poster from a written brief
- Edit one part of an existing image with a text instruction
- Combine several reference images into one composition
Can I use this?
You'll need Python, Diffusers and a GPU-capable runtime · Setup needed
Worth knowing The card’s local example uses a CUDA-capable PyTorch build, with an Apple MPS path also shown, so practical hardware depends on the runtime you choose; it is a 7B model and is not deployed by a Hugging Face Inference Provider. The Qwen Research License Agreement permits non-commercial use only; commercial use requires a separate licence.
Why it's here
Not trending this week
Last seen on the Hugging Face trending list on 21 October 2026. It stays on BuildTube so you can still find it.
- 1,129 people have liked it on Hugging Face.
- It was downloaded 46,769 times in the last 30 days.
Numbers checked on 8 November 2026; not refreshed since.
Behind it
View model pageBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?
Quick questions
- What is Qwen-Image-2.1-Turbo?
- This is the faster checkpoint of Qwen-Image-2.1: you give it a prompt and it generates an image, or you provide a reference image and ask for an edit.
- What can I use Qwen-Image-2.1-Turbo for?
- Make a poster from a written brief; Edit one part of an existing image with a text instruction; Combine several reference images into one composition
- Can I run Qwen-Image-2.1-Turbo on my computer?
- It runs on your own hardware with a model runtime. The card’s local example uses a CUDA-capable PyTorch build, with an Apple MPS path also shown, so practical hardware depends on the runtime you choose; it is a 7B model and is not deployed by a Hugging Face Inference Provider. Python, Diffusers and a GPU-capable runtime.
- What do I need to use Qwen-Image-2.1-Turbo?
- Install a CUDA-compatible PyTorch build or use the documented Apple MPS path, then install the current Diffusers source with Transformers, Accelerate and Pillow. Load the checkpoint with `QwenImage21Pipeline`; a hosted notebook or a compatible local app can provide the runtime instead of a custom Python setup.
- What license does Qwen-Image-2.1-Turbo use?
- The Qwen Research License Agreement permits non-commercial use only; commercial use requires a separate licence.
More from this week
- P-H3-Image-Comfy Make and edit 8-megapixel images in ComfyUI
Turn short text requests into 8-megapixel images and edit pictures on your own graphics card.
- Idol Turn one picture of an object into a 3D Gaussian splat
Idol turns one photo, render or generated image of an object into a 3D Gaussian splat file.
- agent-desktop Let AI agents read and click desktop apps
For developers building computer-use agents: agent-desktop lets an AI agent read and operate desktop apps on Mac, Windows or Linux from a command line.
Explore by topic
Post this
Download share imageFor makers
Made this? Show it’s on BuildTube — add the badge to your README or embed the card on your site.
README badge (Markdown)
[](https://buildtube.app/p/hf-qwen-qwen-image-2-1-turbo/)Embed card (HTML)
<iframe src="https://buildtube.app/embed/hf-qwen-qwen-image-2-1-turbo/" width="560" height="240" style="border:0;max-width:100%" loading="lazy" title="Qwen-Image-2.1-Turbo on BuildTube"></iframe>