Skip to main content
Back to timeline
Vercel NewsSource publication:

Microsoft's MAI-Image-2.7 and Flash arrive on AI Gateway: one key, no Microsoft account, 2048×2048 output

Synopsis

Microsoft has placed MAI-Image-2.7 and MAI-Image-2.7-Flash on AI Gateway in public preview, and both raise the maximum output size to 2048 x 2048 pixels while adding transparent backgrounds for generated and edited images. A developer needs only one AI Gateway API key, with no Microsoft account, to generate images from text or edit existing images. Both models support photorealistic images, text rendering, and targeted edits to products, portraits, and layouts, and up to five reference images can guide edits or combine subjects.

Source-provided article image: MAI-Image-2.7 and MAI-Image-2.7-Flash now available on AI Gateway
Vercel

Interpretation

Microsoft's image models MAI-Image-2.7 and MAI-Image-2.7-Flash can now be called through AI Gateway, so a developer with one API key can generate images from text or edit existing images without holding a Microsoft account. Both models arrive on AI Gateway in public preview, the entry point is a single API key, and the announcement states explicitly that no Microsoft account is needed. Each model carries its own ID, microsoft/mai-image-2.7 and microsoft/mai-image-2.7-flash, and the announcement says setting the AI_GATEWAY_API_KEY environment variable enables calls through AI SDK 6 or later.

Both models raise the maximum output size to 2048 x 2048 pixels and add transparent backgrounds, available for generated images and for edited ones. The output size ceiling and background handling are the capability changes shipped with this launch, and transparent backgrounds cover both the generation and the editing path. In an AI SDK 6 or later call, setting providerOptions.azure.background to transparent yields a transparent PNG, and the example generates a perfume bottle product shot at 2048x2048.

For editing, the original image is passed in as a reference and the change is described in text, with up to five reference images available to guide edits or combine subjects. Editing takes an existing image as input and makes targeted changes to products, portraits, and layouts rather than relying on a text prompt alone. The example uses microsoft/mai-image-2.7-flash to read the generated bottle image, prompts it to place the bottle on a marble counter, and writes out a new image.

Perspective

The launch targets developers using AI Gateway, and both models are in public preview, with scope limited to text-to-image, image editing, photorealistic images, text rendering, and targeted edits to products, portraits, and layouts.

The announcement gives no specific comparison figures for image quality, speed, or cost between the two models; it states only that MAI-Image-2.7 is for final assets where image quality is the priority and Flash for faster, lower-cost prompt exploration and variation.