Midjourney has recently launched a batch of new, more “hands-on” features: an external image editor that lets you upload images for localized edits, an image re-texturing mode, and a more refined V2 moderation system. Below, I’ll walk through how to use these new Midjourney features in practice and what scenarios they’re best suited for.
External Image Editor: From “Generation” to “Editable”
In the past, if you wanted to revise an image in Midjourney, you mostly had to regenerate or rely on lightweight variation tweaks; the core change with the external image editor is: you can upload an image from your computer, then expand it, crop it, repaint it, and even add or replace elements within a specified area.
This editing logic is still driven by the text prompts Midjourney excels at, but it adds control via area selection (masking). As a result, it’s more like turning “localized inpainting” into a visual workflow—ideal for common needs such as fixing composition, filling in backgrounds, or changing props.
How to Use the Midjourney Editor: Three Steps—Upload, Select, Prompt
After entering the editor on Midjourney’s web interface, first upload the image you want to work on, then use the selection tool to box out the area that needs adjustment (for example, a character’s hands, the sky, or where poster text should go). The more precise the selection, the easier it is for Midjourney to make changes according to your intent without affecting parts that shouldn’t be touched.
Next, write in the prompt “what you want it to become,” such as “change the upper-right corner into a neon sign, with stronger night reflections,” then generate comparison results and iterate. Midjourney supports repeated small, incremental edits, which is usually more reliable than trying to cram everything into one ultra-long prompt.


