APIMODELSAPIMODELS
All
Models/MAI Image 2.7 Flash
Microsoft

MAI Image 2.7 Flash API — the production tier of MAI Image 2.7, same request

mai-image-2.7-flash

MAI Image 2.7 Flash is the production tier of Microsoft's MAI Image 2.7: the same text-to-image, editing with up to 5 input images, transparent PNG / WebP and sizes up to 2048×2048, at about 30% of 2.7's per-token rates. It is a public preview with one image per request and no 4K. Call mai-image-2.7-flash at about $0.015 per 1K image; failed requests are not charged.

Text to Image & EditUp to 5 Input ImagesTransparent PNG / WebP1K / 2K up to 2048×2048Per token · ≈$0.015 per 1K image
Per token · text input$2.50 / 1M tokens
Per token · image input$3.58 / 1M tokens
Per token · image output$14.30 / 1M tokens
Typical · 1K image≈ $0.015
Typical · 2K image≈ $0.059
Typical · 1K edit, one input image≈ $0.018
API Docs

Generate Image

Limits: text-to-image, or editing from 1–5 input images · 1 image per request · 1K or 2K, up to 2048×2048, no 4K · billed on the tokens actually used, so the button shows a typical cost · transparent background, JPEG / WebP output and exact sizes are available through the API (PNG here) · a Microsoft public preview that can invent specifications or text, so check the output · not every request succeeds, and failed requests are not charged.

Result

Generated image will appear here

Enter a prompt and click Generate

The production tier of MAI Image 2.7

Microsoft positions Flash for production workloads, and it is the model behind Canvas in MAI Labs. Same request as mai-image-2.7, at about 30% of its per-token rates.

Generate and edit with one model id

A prompt alone, or a prompt plus 1 to 5 input images to recolour, restyle, move an object or combine pictures.

Transparent backgrounds

background: transparent with PNG or WebP output: stickers, icons and product cut-outs in bulk, with no manual background removal.

1K or 2K, up to 2048×2048

Seven aspect ratios (1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16) at 1K or 2K, or an exact size: each side at least 768 px, up to 4,194,304 pixels, rounded down to a multiple of 16. No 4K.

Billed per token, about $0.015 per 1K image

Text input $2.50, image input $3.58, image output $14.30 per 1M tokens. Typical: 1K about $0.015, 2K about $0.059, a 1K edit with one input image about $0.018.

Preview model, not 100%

Microsoft lists it as a public preview. It can invent specifications or labels, so check text and numbers. Not every request succeeds; failed requests are not charged.

MAI Image 2.7 Flash API: the production tier of MAI Image 2.7, same request

MAI Image 2.7 Flash is the Flash tier of Microsoft's MAI Image 2.7, a public preview that Microsoft positions for production workloads and runs behind Canvas in MAI Labs. It takes the same request as MAI Image 2.7: a prompt alone, or a prompt plus 1 to 5 input images, with transparent backgrounds and sizes up to 2048×2048.

On APIMODELS it is one model id, mai-image-2.7-flash, on POST /v1/images/generations, billed per token at about 30% of the 2.7 rates: image output $14.30 per 1M tokens, about $0.015 for a typical 1K image. It returns one image per request, has no 4K, mask or seed, and not every request succeeds; the limits are listed below.

Model ID
mai-image-2.7-flash
Input
Text prompt, plus 1–5 input images to edit (optional)
Output
1 image, 1K or 2K, up to 2048×2048; PNG / JPEG / WebP; transparent background
Aspect ratio
1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16 or an exact size
Price
Per token; about $0.015 per 1K image

MAI Image 2.7 Flash API limits at a glance

What the model accepts and what it does not, before you build on it.

SupportedNot supported
InputA text prompt of up to 32,000 characters, alone (text-to-image) or with 1 to 5 input images to edit, restyle or combineMasks (inpainting); more than 5 input images
Images per requestOneSeveral images in one call (send separate requests)
Sizeaspect_ratio 1:1, 4:3, 3:4, 3:2, 2:3, 16:9 or 9:16 with resolution 1K (about 1 MP, the default) or 2K (each side doubled); or an exact WIDTHxHEIGHT with each side at least 768 px and at most 4,194,304 pixels in total (2048×2048), rounded down to a multiple of 164K; sides under 768 px; other aspect ratios such as 21:9
Edits without a sizeThe model picks a size close to the input image
BackgroundOpaque (default) or transparent; transparent needs PNG or WebP outputTransparent JPEG (JPEG has no alpha channel)
Output formatPNG (default), JPEG or WebP; jpg is accepted as jpegGIF, SVG
Size priorityimage_size (a preset name or {"width", "height"}) wins over size, which wins over aspect_ratio; resolution applies to presets only and is ignored for exact sizes
Other controlsquality is accepted but ignoredseed, negative_prompt, mask and n above 1 return a 400 instead of being silently dropped (results are not repeatable)

Not every MAI Image 2.7 Flash request succeeds

A public preview model with the usual failure modes. You pay only for requests that deliver an image.

Microsoft lists the MAI Image 2.7 family as a public preview, so behaviour, output sizes and quality may change while it is in preview. A request can fail after it is accepted: content review can refuse the prompt or an input image, and a generation can hit a temporary error. A failed request returns the reason and costs nothing. Our one test request on 18 November 2026 (Flash, 1K 16:9 text-to-image) came back as a 1360×768 PNG in about 16 seconds end to end and cost $0.0148; one request is an example, not a success rate or a speed guarantee, so we publish neither.

Check text before you ship it. A hands-on review published on 18 November 2026 found MAI Image 2.7 labels readable in a dense technical illustration, and warns that image models of this generation still invent specifications, dates and part names. Treat model numbers, prices, ingredients and dimensions in a generated image as drafts until a person has read them.

MAI Image 2.7 Flash API pricing

Billed per token on what each request actually uses. The typical costs below are references, not fixed unit prices.

mai-image-2.7mai-image-2.7-flash
Text input, per 1M tokens$7.15$2.50
Image input, per 1M tokens$11.44$3.58
Image output, per 1M tokens$48.62$14.30
Typical 1K text-to-image (about 1,024 output tokens)about $0.05about $0.015
Typical 2K image (each side doubled)about $0.20about $0.059
Typical 1K edit with one input imageabout $0.06about $0.018
Failed requestNot chargedNot charged

100 product shots at 1K

≈ $1.50

100 × about $0.015; failed ones are not charged

One 2K hero banner

≈ $0.059

about 4,096 output tokens × $14.30 per 1M

One 1K edit of a product photo

≈ $0.018

about $0.015 of output plus the input image tokens at $3.58 per 1M

What MAI Image 2.7 Flash is good for, and what it is not

Good fit

Local edits to product photos

Change an object's colour, material or position and keep the rest of the scene. Microsoft leads with this, and with consistency across successive edits.

Good fit

Transparent cut-outs

background: "transparent" with PNG or WebP returns product shots, logos, icons and stickers ready to place on any layout.

Good fit

Catalogue variants at volume

Colourways and backgrounds for many products at about $0.015 per 1K image; send the keepers to mai-image-2.7 if they need more care.

Not a fit

4K, animated GIF or SVG

The largest output is 2048×2048 and the formats are PNG, JPEG and WebP. For 4K, GIF or SVG, use GPT Image 2.6.

Not a fit

Text that must be right without a check

Labels come out readable, but specifications, dates and part names can be invented. Proofread before anything ships.

Not a fit

Masks, many outputs per call, repeatable seeds

There is no mask or seed and one image comes back per request. For up to 10 outputs per call and masks, use GPT Image 2.6.

Call it in one request

POST /v1/images/generations with your apimodels.app API key. Leave out image / images for text-to-image; add them to edit. By default you get a taskId back and poll GET /v1/images/generations?task_id= (or pass a callback_url); with an exact WIDTHxHEIGHT size and no callback_url, the image comes back in the same response. Every field is in the docs.

bash
# Edit: transparent cut-out of a product photo at 1K
curl -X POST https://api.apimodels.app/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mai-image-2.7-flash",
    "prompt": "Cut out the ceramic mug on a transparent background. Keep the mug, its handle and its glaze exactly as they are.",
    "image": "https://example.com/mug.jpg",
    "background": "transparent",
    "output_format": "png",
    "resolution": "1K"
  }'
# → { "code": 200, "data": { "taskId": "...", "state": "pending", "model": "mai-image-2.7-flash" } }

MAI Image 2.7 Flash vs MAI Image 2.7

Same request, different tier. Prices are apimodels.app per-token rates; the leaderboard row is Microsoft's statement.

MAI Image 2.7 FlashMAI Image 2.7
Text input / image input / image output, per 1M tokens$2.50 / $3.58 / $14.30$7.15 / $11.44 / $48.62
Typical 1K · 2K · 1K edit with one input imageabout $0.015 · $0.059 · $0.018about $0.05 · $0.20 · $0.06
Artificial Analysis image editing (per Microsoft, 16 Nov 2026)No figure published#1, 1,215 Elo over 5,478 votes
Microsoft's positioningProduction workloads; runs Canvas in MAI LabsFlagship
Inputs, sizes, transparent backgroundUp to 5 inputs, 1K / 2K up to 2048×2048, PNG / WebPSame

MAI Image 2.7 on apimodels.app →

Sources

Where the third-party facts on this page come from (checked 18 November 2026). Prices and limits are apimodels.app's own.

TL;DR MAI Image 2.7 Flash is a Microsoft image generation model, callable via APIMODELS' unified API (model name `mai-image-2.7-flash`). Pricing: Per token · text input: $2.50 / 1M tokens, Per token · image input: $3.58 / 1M tokens, Per token · image output: $14.30 / 1M tokens, Typical · 1K image: ≈ $0.015, Typical · 2K image: ≈ $0.059, Typical · 1K edit, one input image: ≈ $0.018. One API key for all image / video / LLM / audio models.

About MAI Image 2.7 Flash

MAI Image 2.7 Flash is an Image Generation API provided by Microsoft. MAI Image 2.7 Flash is the Flash tier of MAI Image 2.7, the image model family Microsoft AI (MAI) released on 17 November 2026 at Microsoft Ignite. Microsoft offers it as a public preview, positions it for production workloads, and runs it behind Canvas, the image editor in MAI Labs. On apimodels.app it is one model id, mai-image-2.7-flash, on POST /v1/images/generations, and it takes exactly the same request as mai-image-2.7: a prompt alone for text-to-image, or a prompt plus 1 to 5 input images to recolour, restyle, move an object or combine pictures. It can return a transparent background (background: transparent, with PNG or WebP output), which makes it a practical way to produce stickers, icons and product cut-outs in bulk. Output is PNG by default, or JPEG / WebP. Sizes come from an aspect ratio — 1:1, 4:3, 3:4, 3:2, 2:3, 16:9 or 9:16 — at 1K (about 1 megapixel, the default) or 2K (each side doubled), or from an exact WIDTHxHEIGHT: each side at least 768 px, at most 4,194,304 pixels in total (2048×2048), rounded down to a multiple of 16. There is no 4K, and each request returns one image. Pricing is per token, on the tokens each request actually uses: text input $2.50, image input $3.58 and image output $14.30 per 1M tokens, about 30% of the MAI Image 2.7 rates. A 1K image is about 1,024 output tokens, so as a reference a 1K text-to-image costs about $0.015, a 2K image about $0.059 and a 1K edit with one input image about $0.018. These are typical figures, not a fixed unit price. Failed requests are not charged. Know the limits. The leaderboard result Microsoft published (#1 on Artificial Analysis image editing, 1,215 Elo) belongs to MAI Image 2.7, not to Flash, so for the edits that matter most, compare both on your own prompts. Flash is a preview model, so its behaviour may change. It can still invent specifications, dates and part names in text-heavy images, so check every word and number before you publish. There is no mask, seed or negative prompt. Not every request succeeds: content review can refuse a prompt or an image, and temporary errors happen; a failed request costs nothing. Through APIMODELS platform, you can access this model via a unified API with transparent pay-as-you-go pricing. Current pricing: Per token · text input: $2.50 / 1M tokens, Per token · image input: $3.58 / 1M tokens, Per token · image output: $14.30 / 1M tokens, Typical · 1K image: ≈ $0.015, Typical · 2K image: ≈ $0.059, Typical · 1K edit, one input image: ≈ $0.018.

Related model: MAI Image 2.7 — the tier Microsoft reports #1 on Artificial Analysis image editing — same request, about $0.05 per 1K image

Related model: GPT Image 2.6 Flare — when you need 4K, animated GIF, SVG or up to 10 images per call

Related model: MiniMax H3 Image — when you need up to 16 reference images, 1–10 images per request or 4K

Key Features

  • The production tier of MAI Image 2.7 -- Microsoft positions Flash for production workloads, and it is the model behind Canvas in MAI Labs. Same request as mai-image-2.7, at about 30% of its per-token rates.
  • Generate and edit with one model id -- A prompt alone, or a prompt plus 1 to 5 input images to recolour, restyle, move an object or combine pictures.
  • Transparent backgrounds -- background: transparent with PNG or WebP output: stickers, icons and product cut-outs in bulk, with no manual background removal.
  • 1K or 2K, up to 2048×2048 -- Seven aspect ratios (1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16) at 1K or 2K, or an exact size: each side at least 768 px, up to 4,194,304 pixels, rounded down to a multiple of 16. No 4K.
  • Billed per token, about $0.015 per 1K image -- Text input $2.50, image input $3.58, image output $14.30 per 1M tokens. Typical: 1K about $0.015, 2K about $0.059, a 1K edit with one input image about $0.018.
  • Preview model, not 100% -- Microsoft lists it as a public preview. It can invent specifications or labels, so check text and numbers. Not every request succeeds; failed requests are not charged.

Examples by scenario

Text-to-image, 1K 16:9 · delivered as a 1360×768 PNG · about 16 s end to end, $0.0148 for this request (our one test, 18 Nov 2026) — generated with MAI Image 2.7 Flash
Text-to-image, 1K 16:9 · delivered as a 1360×768 PNG · about 16 s end to end, $0.0148 for this request (our one test, 18 Nov 2026) · aspect_ratio 16:9prompt: A red bicycle leaning against a white brick wall in soft morning light, photograph

Use Cases

Catalogue Variants at Volume

Generate colourways and backgrounds for many products at about $0.015 per 1K image, then send the best ones to mai-image-2.7 if they need the extra care.

Stickers, Icons and Cut-Outs

Transparent PNG or WebP in bulk for apps, slide decks, merchandise and print-on-demand listings.

Interactive Editing in Your App

Let users recolour or move parts of an image on request. Microsoft runs Flash behind its own Canvas editor in MAI Labs.

Drafts Before the Final Render

Explore prompts and layouts with Flash, then render the chosen version with mai-image-2.7: the request is identical apart from the model id.

Social and Ad Creatives

16:9, 9:16 and 1:1 versions at 1K for feeds, stories and A/B tests of an ad.

Thumbnails and Article Images

1K images for blog posts, listings and newsletters where a 2K render would add cost without adding anything visible.

Why APIMODELS

  • Unified API -- One API key to access all models, no need to register on multiple platforms
  • Transparent Pricing -- Pay-as-you-go with per-call pricing published on every model page, ideal for indie developers and startups
  • Instant Access -- Start right after signup; pay by card, PayPal, Alipay or USDT
  • Full Documentation -- Detailed API docs with code examples in cURL, Python, and Node.js

Frequently Asked Questions

How to use MAI Image 2.7 Flash API?

Sign up at APIMODELS, get your API key, and call our unified API endpoint. We provide detailed API documentation with code examples in cURL, Python, and Node.js.

What is the difference between APIMODELS and the official Microsoft API?

APIMODELS offers the same MAI Image 2.7 Flash model through our aggregation platform. We provide a unified API interface so you do not need separate accounts for each provider - one API key to access all models.

How much does MAI Image 2.7 Flash cost?

mai-image-2.7-flash is billed per token on apimodels.app, on the tokens each request actually uses: text input $2.50, image input $3.58 and image output $14.30 per 1M tokens. A 1K image is about 1,024 output tokens, so a 1K text-to-image costs about $0.015, a 2K image about $0.059 and a 1K edit with one input image about $0.018. These are typical costs, not a fixed unit price; failed requests are not charged.

What is the difference between MAI Image 2.7 Flash and MAI Image 2.7?

They take the same request: text-to-image or editing with up to 5 input images, transparent PNG / WebP, 1K or 2K up to 2048×2048, one image per request. Flash is the tier Microsoft positions for production workloads and the model behind Canvas in MAI Labs; its per-token rates are about 30% of 2.7's (image output $14.30 against $48.62 per 1M tokens). The leaderboard result Microsoft published (#1 on Artificial Analysis image editing, 1,215 Elo) is for MAI Image 2.7, not Flash, so test both on your hardest edits.

Microsoft AI: Introducing MAI Image 2.7 ↗

Can MAI Image 2.7 Flash edit images?

Yes. Send the prompt with the picture as image, or up to 5 pictures as images, to POST https://api.apimodels.app/v1/images/generations with model mai-image-2.7-flash. Say what to change and what to keep, for example "turn the label teal and keep the bottle, light and background as they are". Each input image adds image-input tokens at $3.58 per 1M; a 1K edit with one input image is about $0.018 in total.

Does MAI Image 2.7 Flash support transparent backgrounds?

Yes, the same way as MAI Image 2.7: background transparent with output_format png or webp (JPEG has no alpha channel). At about $0.015 per 1K image, that makes Flash a practical choice for large batches of stickers, icons and product cut-outs.

What sizes does MAI Image 2.7 Flash support?

An aspect ratio — 1:1, 4:3, 3:4, 3:2, 2:3, 16:9 or 9:16 — at resolution 1K (about 1 megapixel, the default) or 2K (each side doubled), or an exact WIDTHxHEIGHT with at least 768 px on each side and at most 4,194,304 pixels in total (2048×2048), rounded down to a multiple of 16. There is no 4K. Edits without a size come back close to the input image's size.

What are the limits of MAI Image 2.7 Flash, and what happens when a request fails?

Flash is a public preview, so its behaviour may change. One image per request, up to 5 input images, no mask, seed or negative prompt, and no 4K. In text-heavy images it can invent specifications, dates or part names, so proofread anything a customer will read. Not every request succeeds: content review can refuse a prompt or an image, and temporary errors happen. A failed request returns the reason and is not charged.

CIAOPS hands-on review, 18 Nov 2026 ↗

How does MAI Image 2.7 Flash compare with other models in the Image Generation category?

On APIMODELS, MAI Image 2.7 Flash runs alongside 60+ models on one API key and one balance, so choosing is about fit, not lock-in. It supports Text to Image & Edit, Up to 5 Input Images, Transparent PNG / WebP, 1K / 2K up to 2048×2048, Per token · ≈$0.015 per 1K image, and you can weigh it on price and capability against other models in the Image Generation category, then switch by changing a single model-name string — no new account or integration. Browse every model in that category with live pricing at apimodels.app/models.

What can MAI Image 2.7 Flash do?

MAI Image 2.7 Flash supports: Text to Image & Edit, Up to 5 Input Images, Transparent PNG / WebP, 1K / 2K up to 2048×2048, Per token · ≈$0.015 per 1K image. See the APIMODELS docs for full parameters and call examples.

Can I use the MAI Image 2.7 Flash API from any country, including mainland China?

Yes. APIMODELS does not restrict access by country: the site and the API answer requests from anywhere, and apimodels.app and api.apimodels.app are reachable directly from mainland China without a proxy. You call MAI Image 2.7 Flash with one APIMODELS key, so you do not need a separate account with the model's developer.

What payment methods are supported? Can I pay with PayPal or crypto?

Yes to both. You can top up with a card (Visa, Mastercard and others, through Stripe), PayPal, Alipay, or USDT on TRON (TRC20) or BNB Smart Chain (BEP20). Card, PayPal and Alipay top-ups are credited immediately; USDT is credited once the transfer confirms on-chain. Auto top-up is optional, and a $50 top-up adds a $1 bonus ($3 on a $100 top-up). The balance is in USD and billed per use, with no subscription.

Related models