As a popular tool in AI image generation, Midjourney has recently rolled out several practical updates. From improved character body consistency to enhanced detail in larger models, these feature changes directly impact output quality. This article focuses on the core changes in the new features, helping creators quickly master key techniques to elevate image expressiveness.
Character Consistency: Say Goodbye to Distorted Limbs
In the latest update, Midjourney has significantly optimized the generation logic for human body parts. The consistency of elements like arms, legs, and palms has been markedly improved, and common issues such as incorrect finger counts or twisted arms are now greatly reduced. When using the new version, inputting prompts involving multi-character interactions—for example, "two people shaking hands"—produces more natural and coordinated results, with noticeably enhanced overall visual unity.
Details & Textures: Quality Reaches a New Level
The new version has put serious effort into eliminating pixel artifacts and enhancing skin texture. Local details like eyes, lips, and strands of hair are rendered with higher precision, and small distant objects are no longer blurry. The newly added --q 2 mode, while extending generation time by about 25%, delivers richer texture details—ideal for scenes that demand high visual quality.

