Microsoft's MAI-Image-2.7-Preview takes #2 on image editing arena, closing in on GPT Image 2.5
ArtificialAnlys · x · 2026-10-27
Microsoft AI released MAI-Image-2.7-Preview, its newest image editing model, which ranks #2 on the Artificial Analysis Image Editing v2.1 leaderboard, just behind OpenAI's GPT Image 2.5 Sunburst (max) and a clear step up from MAI-Image-2.6 (August).
- Scores higher than 2.6 on all 7 measured editing actions, ranking #1 on 4: Scene & Style Edit, Text or Symbol Edits, Reasoning-Based Edit, and Enhancement & Restoration
- Biggest Elo gains: Composition & Framing (+57), Text or Symbol Edits (+49), Scene & Style Edit (+48)
- Ranks #1 on 4 of 10 use cases: Live-Action Film, Consumer, Frontier, and UI/UX Design
- Available now in MAI Playground for editing only; text-to-image coming soon
Related event: Microsoft's MAI-Image-2.7-Preview Climbs Image Editing Leaderboards(9 posts)→
More from Models
- Grim Test finds AI shows no mercy for humans even while causing suffering — hollee-o · 2026-10-28
- LMArena launches Code Arena WebDev leaderboard for model coding comparisons — arena · 2026-10-28
- DeepMind unveils Genie 4, its most realistic and interactive world model yet — GoogleDeepMind · 2026-10-28
- Local coding model benchmarks should measure the whole setup, not just tokens/sec — palsc_ · 2026-10-28
- Dev builds AI agent controlling macOS in a Daytona sandbox, benchmarked across 5 LLMs — jamesmurdza · 2026-10-28
- 693 LLM inference throughput benchmark runs across 228 models land on LocalMaxxing — TheZachMueller · 2026-10-27