Microsoft MAI-Image-2.5 for photorealistic and artistic text-to-image generation.
Added Jun 3, 2026
Approx. Price
$0.005 per image
Model Type
text-to-image
Settings
Generation controls available for this model.
Images Per Run
Up to 1
Output images
Input Images
Up to 1
Reference/edit images accepted • Route max 30 MB
Output Sizes
Aspect Ratio
Default
auto
Number of Images
Default
1
Resolution
Default
auto
Options (8)
auto (Auto), 1:1 (Square), 16:9 (Landscape), 9:16 (Portrait) +4 more
Benchmarks
Benchmarks
Human preference benchmarks sourced from Artificial Analysis.
Text to Image
#5 / 147
ELO
1303.0
Appearances
15,678
95% CI
-9/9
Image Editing
#2 / 69
ELO
1256.0
Appearances
11,732
95% CI
-8/8
Release Date 2026-06 · Matched as MAI-Image-2.5
Artificial Analysis APIRelated image models
Compare MAI-Image-2.5 with similar models from the same provider or model family.
Krea 2 Turbo
wavespeed-ai/krea-v2/turboKrea 2 Turbo is a fast 12.9B open-weights image model with native 1K and 2K generation plus optional image-to-image guidance.
Krea 2 Turbo LoRA
wavespeed-ai/krea-v2/turbo-loraKrea 2 Turbo LoRA adds support for up to three custom LoRA adapters while retaining native 1K/2K and image-to-image generation.
Grok Imagine Image 2.0 Edit
xai/grok-imagine-image/v2.0/editGrok Imagine Image 2.0 Edit applies prompt-guided changes using up to three input images, with flexible aspect ratios, 1K or 2K output, and selectable low or medium quality. Requests rejected under SpaceXAI's terms may still incur a provider charge.
Grok Imagine Image 2.0
xai/grok-imagine-image/v2.0/text-to-imageGrok Imagine Image 2.0 generates detailed images with strong prompt following, flexible aspect ratios, 1K or 2K output, and selectable low or medium quality. Requests rejected under SpaceXAI's terms may still incur a provider charge.
Qwen Image 3 Pro
qwen-image-3-proAlibaba's professional Qwen Image 3 model for high-quality 1K or 2K text-to-image generation and precise edits with up to 3 reference images.
Higgsfield Soul Image-to-Image
higgsfield/soul/image-to-imageHiggsfield Soul transforms a reference image into a stylized, realistic scene while preserving character identity. Choose from medium or high quality, a wide range of creative presets, and adjustable style strength.