MidjourneyIntroduces

作者：大自然的搬运工 · 2026-06-16 12:17:54

Midjourney has launched the alpha version of its V7 image model, marking its first major architecture update in nearly a year. The release introduces a default "personalization" feature that learns individual aesthetic preferences, alongside a faster, lower-cost "Draft Mode" for rapid iteration. The V7 model boasts significant improvements in image quality, particularly in rendering hands, bodies, and natural textures, while offering more coherent prompt interpretation. A new conversational interface allows users to refine images through natural language commands, moving beyond complex parameter syntax. As an alpha release, current limitations include the absence of upscaling, retexturing, and some advanced V6 features. The company plans to integrate these capabilities in the coming months, with a V7 character and object reference system expected in the near future.