Skip to main content
Core.Today
|
MicrosoftMediumUltra

Microsoft MAI-Image 2.5 Pro

Microsoft's highest-fidelity image model for production-grade text-to-image generation via Fal.AI. Built for hero imagery, detailed compositions, precise text rendering, photorealism, stylized illustration, commercial design, and visually rich concept work.

400 credits
per image
High-fidelity output tuned for production and commercial use
Precise in-image text rendering
Generate up to 4 images per call
Multiple aspect ratios, including an auto mode that lets the model choose
Output as JPEG, PNG, or WebP
Optional data-URI sync mode for outputs that skip request history

Run it right now

Test this model instantly in the Console Playground โ€” no code required

Sign in to try

Use with AI Assistant

Copy usage instructions for Claude, ChatGPT, or other AI

Quick Start

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "microsoft/mai-image-2.5-pro",
  "input": {
    "prompt": "A photorealistic concept-art poster of a university campus at sunset, cinematic lighting, a banner reading \"LOUISVILLE\" in bold serif type.",
    "aspect_ratio": "auto",
    "output_format": "png",
    "num_images": 1
  }
}'

Parameters

ParameterTypeRequiredDefaultDescription
promptstringYes-The text prompt to generate an image from.
num_imagesintegerNo1The number of images to generate.
aspect_ratiostringNoautoThe aspect ratio of the generated image. Use "auto" to let the model decide based on the prompt.
auto1:14:33:416:99:163:22:3
output_formatstringNopngThe format of the generated image.
jpegpngwebp
sync_modebooleanNofalseIf `True`, the media will be returned as a data URI and the output data won't be available in the request history.

Common Parameters

Common parameters used when calling POST /v1/predictions.

ParameterTypeRequiredDefaultDescription
modelstringYes-Model identifier
inputobjectYes-Object containing the model-specific parameters from the table above
output_folderstringNo-Folder path for output files (max 256 chars, '..' not allowed)
webhook_urlstringNo-Webhook URL to call on completion
is_publicbooleanNofalseIf true, output files are also available via permanent public URLs

Examples

Hero poster with precise banner text

Generate a cinematic marketing poster where a specific headline must render legibly in the image.

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "microsoft/mai-image-2.5-pro",
  "input": {
    "prompt": "A photorealistic concept-art poster of a university campus at sunset, cinematic lighting, a banner reading \"LOUISVILLE\" in bold serif type.",
    "aspect_ratio": "auto",
    "output_format": "png",
    "num_images": 1
  }
}'

Batch variants for social formats

Generate multiple images at once in a fixed vertical aspect ratio for stories/reels-style placements.

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "microsoft/mai-image-2.5-pro",
  "input": {
    "prompt": "A stylized illustration of a coffee cup on a marble counter, warm morning light, minimalist commercial design.",
    "aspect_ratio": "9:16",
    "output_format": "webp",
    "num_images": 4
  }
}'

Tips & Best Practices

1Leave aspect_ratio at auto to let the model choose based on the prompt, or set an explicit ratio when you need consistent output across a batch.
2For in-image text, quote the exact wording in the prompt (e.g. a banner reading "...") to get more accurate rendering.
3Use sync_mode only when you need the image immediately inline and don't need it to persist in request history.
4Raise num_images to generate several compositions from one prompt and pick the best result.

Use Cases

Hero imagery and banners for marketing campaigns and landing pages
Posters or graphics that require precise, legible text rendering
Photorealistic product or lifestyle concept imagery
Stylized illustration for commercial design work
Batch-generating multiple aspect-ratio variants of the same concept