Microsoft adds Flash tier to MAI-Image-2.6
Microsoft AI has made MAI-Image-2.6 available to developers in Foundry and added a faster Flash variant, pitching its image models at teams that need near-top quality at production speed and cost.
Microsoft AI announced that MAI-Image-2.6, which it calls its strongest image model yet, is now available to developers in Microsoft Foundry. It also introduced MAI-Image-2.6-Flash, a variant built for latency-sensitive, high-throughput workloads.
The company cites third-party leaderboards for the main model. As of September 4, 2026, MAI-Image-2.6 ranked No. 2 for both text-to-image and image editing on Arena. On Artificial Analysis it ranked No. 2 for text-to-image and No. 1 for image editing.
The Flash model's figures come from Microsoft itself. The company says Flash generates images 2.8x faster than OpenAI's GPT-Image-2-Medium and delivers "72% greater efficiency," but did not say how efficiency was measured. Microsoft also claims MAI-Image-2.6 offers "the best price-per-Elo performance in the world." The announcement gives no per-image or per-token prices for either model.
Both models support:
- Multi-reference editing, which combines people, products, styles and scenes from several input images
- Web grounding, which pulls current information from the web into generated visuals
- Dynamic aspect ratios, with output up to 1.5K resolution
Microsoft frames the pair as a choice between maximum precision with MAI-Image-2.6 and production speed with Flash. The post does not give Flash's own leaderboard rankings, its availability outside Foundry, or details of the benchmark setup behind the speed comparison.