MAI-Image-2.6 Edit
Microsoft AI's flagship image-to-image editing model, combining surgical instruction-driven edits with multi-reference composition across up to five images, ranked No. 1 for image editing on Artificial Analysis.
What it does
MAI-Image-2.6 Edit, in practice.
MAI-Image-2.6 is Microsoft AI's flagship image model, built for high-quality generation and precise, controllable editing. This endpoint exposes its image-to-image path: you supply 1–5 reference images plus a natural language instruction, and the model returns an edited or newly composed image. MAI-Image-2.6 is a diffusion-based generative model that progressively refines an image from the prompt and the supplied references. It handles both surgical edits — remove, replace, recolor, reposition, inpaint, fix text, clean up artifacts — and multi-reference composition, where people, products, styles, and scenes from several different images are brought together into one coherent result. Compare
- Multi-reference composition — Accepts up to 5 reference images in a single request. A locked creative — a product, a character, a brand mark — stays consistent as you revise, recompose, and resize it. This is the defining capability of the 2.6 editing path.
- Surgical object editing — Remove, replace, recolor, or reposition specific elements without disturbing the rest of the frame.
- Inpainting and artifact cleanup — Fill missing regions, erase unwanted content, and remove blur, noise, and compression artifacts.
- In-image text updates — Rewrite signage, labels, packaging copy, and UI strings with legible, correctly spelled results. Microsoft measured a +91 Elo gain on text rendering over MAI-Image-2.5.
- Identity and layout preservation — Keeps faces, subject identity, and composition intact across edits.
- Style and layout transfer — References can act as content, layout, or style guides rather than only as the canvas being edited.
Run MAI-Image-2.6 Edit
from $0.13/imageParameters
What you can set.
| Parameter | What it does | Options |
|---|---|---|
prompt | Text prompt describing the edit to perform on the input images. Maximum context length: 32,000 tokens. | |
reference_images | The images to edit, also used as content, layout or style references. Accepts 1 to 5 images. Must be in JPEG or PNG format. Supports both a public URL and a base64-encoded image for each item. The total pixel size of each image should be le | |
size | Controls the aspect ratio of the edited image. "default" preserves the aspect ratio of the input image, and "auto" lets the model evaluate the prompt and the provided image inputs and select the output aspect ratio it determines is best sui | default auto |
enable_web_search | If enabled, the model will use web search to ground the generation with real-time information. This can improve accuracy for prompts involving real-world entities, places, or events. | default |
Sample prompt
The prompt behind the sample.
Design a promotional advertisement image for a watch featuring a snow-capped mountain theme and incorporating text.
size: defaultenable_web_search: enable_base64_output: enable_sync_mode: FAQ
Short answers.
How much does MAI-Image-2.6 Edit cost?
Pricing starts at $0.13 per image at the base resolution; higher resolutions and longer durations cost more. Usage is billed per request from your balance — no subscription.
Does MAI-Image-2.6 Edit run uncensored here?
No. Microsoft applies its own content policy to this model regardless of where it is called from. For a route without an extra platform filter use Uncensored MiniMax H3; this page lists MAI-Image-2.6 Edit for everything else it does well.
What does MAI-Image-2.6 Edit take as input?
It is a image editing model. Microsoft AI's flagship image-to-image editing model, combining surgical instruction-driven edits with multi-reference composition across up to five images, ranked No. 1 for image editing on Artificial Analysis.
How do I use it?
Enter an invite code to open Chat with the model loaded, or join the waitlist. We open seats in batches and track which models are requested most.
Related