Qwen-Image-Edit-2511.
Qwen-Image-Edit-2511 is an open-source image model from Alibaba (Qwen team). Windpaint doesn't serve it yet, so there's no model id or price. Request early access to start with the live models.
- Status
- Coming soon
- Task
- Image editing
- Input
- One or more images and an instruction
- Output
- Edited image
- Developer
- Alibaba (Qwen team)
- Parameters
- 20B
- Licence
- Apache-2.0
- Released
- December 2025
What the model does.
Paraphrased from the official model card.
Qwen-Image-Edit is the editing version of Qwen-Image. You give it an image and a written instruction, and it changes the image: swap an object, change a style, move a person to a new scene, or edit text in the picture while keeping its font.
The 2511 release builds on Edit-2509, which added edits that combine several input images. Its card says 2511 drifts less from the source image, keeps characters more consistent across edits, and has selected community LoRA effects built in.
Coming soon to Windpaint.
Qwen-Image-Edit-2511 isn't served, so it has no model id, no price and no request example here. When it goes live, this page gets its tiers, prices and request fields like the live models have.
Want this model? Request early access. We're adding models as we add capacity.
Live for images today: Z-Image Turbo (text to image) and FLUX.2 [klein] 4B (text to image). See Image generation for what the API does now.
Is Qwen-Image-Edit-2511 licensed for commercial use?
Licence: Apache-2.0.
Yes. Apache-2.0 allows commercial use, including running the model behind a paid API, with no revenue or user caps.
- Apache-2.0 (card metadata and the card's License Agreement section). Qwen-Image-Edit-2511 model card
This is our reading of the licence, not legal advice. Read the licence text before you rely on it.
Qwen-Image-Edit-2511 hardware requirements.
Published by Alibaba (Qwen team) for running the weights yourself. These are the developer's figures, not Windpaint's.
- Published VRAM
- Not stated
- Applies to
- No VRAM figure on the card. bf16 on CUDA.
- The card loads
QwenImageEditPlusPipelinein bfloat16 on CUDA. It gives no VRAM figure. Qwen-Image-Edit-2511 model card - The official README's community section says DiffSynth-Studio (from ModelScope, a community tool) supports Qwen-Image with layer-by-layer offload (inference within 4GB VRAM) and FP8 quantization. It doesn't say whether this covers Edit-2511. Qwen-Image repo
Recommended settings for Qwen-Image-Edit-2511.
Steps, guidance and defaults from the developer's card and code, for running the weights yourself.
true_cfg_scale=4.0,negative_prompt=" ",num_inference_steps=40,guidance_scale=1.0,num_images_per_prompt=1. Qwen-Image-Edit-2511 model card- The image input is a list. The card's example passes two images in one call. Qwen-Image-Edit-2511 model card
Qwen-Image-Edit-2511 prompt guide.
From Alibaba (Qwen team)'s own prompt guide, prompt tooling and examples.
- For multi-image edits, the card's example says where each subject goes: "The magician bear is on the left, the alchemist bear is on the right, facing each other in the central park square." Qwen-Image-Edit-2511 model card
- For image editing, the repo recommends its official prompt enhancer (
polish_edit_prompt) "for enhanced stability". It uses Qwen-VL-Max and takes the input image as well as the prompt. Qwen-Image repo - The README's 2511 example puts the text to show in quotes: a TV screen that reads “阿里巴巴”. Qwen-Image repo
Qwen-Image-Edit-2511 limitations.
As the developer states them.
- The card for Edit-2509, the release that 2511 builds on, says multi-image editing currently works best with 1 to 3 input images. The 2511 card doesn't restate a limit. Qwen-Image-Edit-2509 model card
Questions.
Can I call Qwen-Image-Edit-2511 on Windpaint today?
No. Qwen-Image-Edit-2511 isn't served yet, so it has no model id, no price and no entry in the catalog endpoint. For images today, z-image-turbo and flux-2-klein-4b are live. There's no date for Qwen-Image-Edit-2511.
What changed in Qwen-Image-Edit-2511 vs 2509?
The card lists five areas: image drift, character and multi-person consistency, selected popular community LoRAs built into the base model (such as lighting enhancement and new viewpoints), industrial design generation, and geometric reasoning (such as drawing auxiliary construction lines).
Can Qwen-Image-Edit-2511 edit with multiple images?
Yes, it takes a list of input images. The Edit-2509 card, from the release that added multi-image input, says 1 to 3 images work best.
What licence is Qwen-Image-Edit-2511 under?
Apache-2.0, which allows commercial use.
What negative prompt does Qwen-Image-Edit-2511 use?
The card's example passes a single space as the negative prompt, with true_cfg_scale 4.0 and 40 steps.
Is there a prompt enhancer for Qwen-Image-Edit-2511?
Yes. The repo has an official edit-prompt enhancer (polish_edit_prompt) that uses Qwen-VL-Max and takes the input image.
Is Qwen-Image-Edit-2511 open source?
Yes. Its weights are licensed Apache-2.0, an OSI-approved open-source licence.
How do I ask for Qwen-Image-Edit-2511?
Request early access at app.windpaint.ai/register. We're letting people in gradually while we add models and capacity.