Image
Describe it
Write what you want to see in the box at the bottom. One or two sentences is enough. Memo Magic fills in the detail for you. Switch to Simple prompt to send your words exactly as written. See Prompt modes.Add a reference
Press + to attach a logo, a product photo, a saved inspiration or an earlier result. The model follows it. This step is optional, and some models take no references: the button tells you when that is the case.Pick the model and the shape
Choose the model, the aspect ratio, the quality and how many images you want. The number beside Generate is what this run costs in credits. It changes as you change the settings.Generate, then keep what works
Results collect on the page. On any result you can Retry, Reuse parameters, download it, or add it to a collection.Video
Describe the clip
Write what happens, in the order it happens. Camera moves and mood help.Start from an image
Add a Start frame to animate a picture you already have. Some models also take an End frame or reference images. This step is optional.Pick the model, the length and the shape
Choose the model, the duration, the aspect ratio and the quality. Longer and sharper clips cost more. The number beside Generate is the price of this run.Wait, then keep what works
A clip takes longer than an image. It appears on the page when it is ready, where you can retry it, reuse its settings or download it.Models that run on Higgsfield
A few models run on Higgsfield, a second provider: Soul 2 for images, Seedance 2.5 · Higgsfield and Seedance 2.0 · Higgsfield for video.- Soul 2 makes portrait and fashion images from text at the lowest price in Memo. It takes no references.
- Seedance via Higgsfield keeps photoreal character faces. The other Seedance entries refuse them.