Is MAI Image 2.6 available through an API?

The MAI Image 2.6 integration uses POST /api/v1/images/generations with model mai-image-2.6. Submission returns a task ID; retrieve the finished output by polling the tasks endpoint.

How is MAI Image 2.6 priced?

MAI Image 2.6 billing depends on text input, reference images, and generated pixels. The table shows square-image estimates without references. Credits are reserved at submission and reconciled after completion; the final charge can differ. Failed tasks are refunded.

Is MAI Image 2.6 free on reAPI?

MAI Image 2.6 is a paid, pay-as-you-go model; no monthly subscription is required. Review the current estimate before generating. Polling an existing task does not create another image request.

How many references does MAI Image 2.6 support?

MAI Image 2.6 accepts up to five public HTTP(S) image URLs. Keep their order stable and explain each image’s role. The prompt remains required when references are supplied.

Can MAI Image 2.6 generate 4K images?

MAI Image 2.6 supports 1K and 2K tiers, not 4K. Explicit dimensions require both sides to be at least 768 pixels and a total area no greater than 2,359,296 pixels.

Why did MAI Image 2.6 return a different size?

MAI Image 2.6 reference edits use model-selected dimensions and ignore the sizing controls. For text-only requests, explicit width and height take priority over pixel size, then ratio plus resolution. Nonmultiples of 32 are rounded down.

Does MAI Image 2.6 support automatic framing?

Yes. Use size auto or enable auto_aspect_ratio to ask MAI Image 2.6 to infer a suitable frame from the prompt. Output size and the final cost can vary.

What does MAI Image 2.6 web grounding do?

The MAI Image 2.6 web_grounding switch allows web context during generation and defaults to false. Treat any depicted facts, labels, and details as material to review, not as verified information.

Is MAI Image 2.6 the same as Flash?

No. Microsoft presents Flash as a separate model focused on speed and cost. This page and the model ID mai-image-2.6 refer to standard MAI Image 2.6; they do not select Flash.

Can I use masks or a quality setting?

This MAI Image 2.6 endpoint does not expose mask_url, quality, style, background, output_format, or response_format. Use prompts and reference images for edits. Unsupported request fields are rejected.

How do I get an API key for MAI Image 2.6?

Create a reAPI account, open Settings → API Keys, and create a key. Keep it on your server and send it as a Bearer token with MAI Image 2.6 requests to the images endpoint.

MAI Image 2.6 — Create Images With Clear Direction

Turn a brief into a product scene, refine an image with references, or explore a new visual layout. MAI Image 2.6 combines image creation, editing, and optional web context in one workflow.

Highlights: Text-to-image · 5 references · 1K and 2K
Best for: Product scenes, reference edits, visual layouts
Input: Text and optional reference images
Output: One PNG image per request

Input

Prompt*

Required for generation and editing. The documented context limit is approximately 32,000 tokens.

Reference images

Drag images here or click to upload

0/5 files

Public HTTP(S) URLs only. With references, the model chooses output size and ignores the size controls.

Size or aspect ratio

Use auto, a ratio between 1:4 and 4:1, or dimensions such as 1536x1024. Applies to text-to-image.

Resolution

MAI Image 2.6 supports 1K and 2K, not 4K. Explicit pixels override this setting.

Automatic framing

Off by default. Equivalent to requesting size auto.

Web grounding

Off by default. MAI Image 2.6 still needs a clear prompt; review factual details in the result.

Estimated cost~$0.03838 credits

Result

Try one of these prompts

Pricing

Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.

First test cost

$0.03838 credits

One sample at the cheapest tier (estimated image).

Testing budget guide

Add credits

$10

≈ 267 tests

$50

≈ 1338 tests

$100

≈ 2676 tests

1K · no references · estimate: $0.038; Price · estimated image
2K · no references · estimate: $0.085; Price · estimated image

What is MAI Image 2.6?

MAI Image 2.6 is Microsoft AI’s image generation and editing model. Describe a new scene or supply reference images and explain the changes you want. MAI Image 2.6 supports up to five references, optional web grounding, and flexible framing for text-to-image work. On reAPI, MAI Image 2.6 uses the images endpoint and returns an asynchronous task ID. The completed task contains a PNG image. Start with a simple composition, inspect the result, and add instructions when the visual direction needs refinement. This page covers the standard model; MAI Image 2.6 Flash is a separate product.

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.

MAI Image 2.6 scenario illustration with an ivory teapot and product composition references

Create a product scene with MAI Image 2.6

Turn a written brief into a studio photograph, a material study, or a campaign concept. Tell MAI Image 2.6 the subject, surface, lighting, and camera view. A ceramic teapot on a pale blue table gives MAI Image 2.6 a concrete starting point; a vague request for a beautiful product image leaves more decisions open. Give MAI Image 2.6 references when the product already exists. Check its silhouette, packaging, and labels before treating a generated concept as a faithful product representation.

Read the MAI Image 2.6 API guide

MAI Image 2.6 reference editing concept with a blue chair, room image, and material swatches

Guide MAI Image 2.6 with reference images

Combine a room photograph, a furniture reference, and a color study in one editing request. Tell MAI Image 2.6 which image supplies the scene and which supplies the object. Keep the reference order stable and tell MAI Image 2.6 which details should remain unchanged. MAI Image 2.6 chooses the output framing for reference edits, so fixed dimensions are not a delivery guarantee. Inspect subject identity, edges, reflections, and surrounding details before selecting a result for your project.

MAI Image 2.6 layout illustration with a museum poster and a square sculpture composition

Explore a layout with MAI Image 2.6

Develop an exhibition poster, a social visual, or an editorial illustration with space for a headline. Ask MAI Image 2.6 for a specific hierarchy, restrained colors, and a clear focal point. For a new MAI Image 2.6 image, choose a square, portrait, or landscape frame, or supply valid pixel dimensions. Put short required wording in quotation marks. MAI Image 2.6 can generate lettering, but proofread it at the final viewing size and add precision typography in your design tool when needed.

Why reAPI for MAI Image 2.6

MAI Image 2.6 starts from your brief

Brief MAI Image 2.6 in ordinary language to describe materials, mood, and composition. Give MAI Image 2.6 the important constraints first, then inspect whether the result follows them before adding more detail.

MAI Image 2.6 accepts visual context

Bring up to five references into one MAI Image 2.6 request. Clear roles help MAI Image 2.6 distinguish the subject to preserve from the style or setting to change.

MAI Image 2.6 makes framing explicit

Set the MAI Image 2.6 frame with a ratio or explicit dimensions. Review the estimate before submitting; MAI Image 2.6 settles the final charge after generation, so the initial estimate is not a fixed price.

MAI Image 2.6 vs MAI Image 2.6 Flash

Microsoft describes Flash as the faster, lower-cost variant. MAI Image 2.6 is the standard model integrated on this page; Flash is a separate model and is not selected by its request ID.

Model choice

MAI Image 2.6

MAI Image 2.6 Flash

Microsoft positioning

High-quality generation and controllable editing.

Faster generation and lower cost.

Documented task types

Text-to-image generation and reference editing.

reAPI selection here

Use model mai-image-2.6.

Not selected or integrated on this page.

Choosing a model

Evaluate when detailed output is the priority.

Evaluate separately when speed and cost are priorities.

Microsoft product positioning, reviewed October 9, 2026; this is not a measured quality or speed comparison. Scenario illustrations explain workflows and are not benchmark outputs.

Start using MAI Image 2.6

step 01
Get your MAI Image 2.6 key
Sign in and create a key under Settings → API Keys. Keep it on your server and use your reAPI balance for MAI Image 2.6 requests.
Open
step 02
Prepare a MAI Image 2.6 brief
Describe a concrete subject for MAI Image 2.6 and choose a frame. Add references only when they contribute useful context. Check the estimate and the reference-edit sizing note.
Open
step 03
Review your MAI Image 2.6 result
Submit model mai-image-2.6 to the images endpoint. Poll the returned task ID until completed or failed. Inspect the MAI Image 2.6 output before downloading or sharing it.
Open

docs/api/mai-image-2-6

API reference

Drop-in code and the full parameter table.

View full reference

curl https://reapi.ai/api/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mai-image-2.6","prompt":"A matte ivory ceramic teapot on a pale blue table, soft daylight, product photography.","size":"1:1","resolution":"1K"}'

Parameter validation failed

Rejected synchronously with the failing field named. Check enum values, ranges, and URL fields — nothing is charged.

Authentication or balance issue

401 means a missing or invalid Bearer key; 402 means the reserve exceeds your balance. Manage keys and credits in the console.

Content or material rejected

Reference material and outputs pass automated moderation. Rejected tasks fail with a clear error and a full refund.

Task failed or timed out

A task that reaches a failed state is never charged — the reserve refunds automatically. Keep the task id and retry when ready.

Frequently asked questions

Common questions about this model.

MAI Image 2.6 generates and edits still images from text and optional references. Each request returns one PNG. Product concepts, reference-guided scenes, and poster layouts are useful starting points.