MAI-Image-1
Microsoft's first in-house text-to-image model — top-10 LMArena debut, photoreal focus.
About
MAI-Image-1 is Microsoft AI's first text-to-image generator designed and developed in-house, announced in early November 2025 and publicly testable from November 4, 2025 — another step in Microsoft's push for a self-owned model stack alongside MAI-1-preview and MAI-Voice-1. Microsoft says it solicited feedback from creative professionals during training to avoid repetitive or generically-styled outputs, targeting photorealistic imagery with strong lighting (bounce light, reflections) and landscapes, plus a speed-plus-quality pitch for fast iteration.
It debuted in the top 10 of the LMArena image arena — an independent placement, though announced by Microsoft itself. Access is consumer-facing: LMArena testing, Bing Image Creator (alongside DALL-E 3 and GPT-4o, which Microsoft encourages users to compare), and Copilot Audio Expressions' Story Mode.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Independent placement (LMArena): top-10 image arena at launch, per Microsoft's announcement of the result.
Provider-reported: photorealism and speed advantages are qualitative claims.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Announcement and LMArena top-10 placement confirmed across independent press (Windows Central, ZDNet) with consistent dates. Docked for: (1) the arena placement is reported by Microsoft itself rather than observed independently in coverage; (2) quality claims (lighting, photorealism) are qualitative and provider-framed; (3) no API or weights — consumer surface only.