LEARN

Every feature, in plain steps.

Short and practical — each guide is the exact click path. Inside the editor, hover any button for a full explanation, and press ? for keyboard shortcuts.

Your first layered PSDSeparate a subject from its backgroundDecompose a whole scene at onceTurn flat text into editable text layersFix a selection by handBatch: 10–200 images in one job

Your first layered PSD (2 minutes)

The core loop of Layersmith — everything else builds on this.

  1. Open your image. Drag any JPG, PNG, or WebP onto the editor, or click the drop zone. Free tools run entirely in your browser — the image never uploads.
  2. Click the object with AI grab. AI grab is selected by default (key A). Click once anywhere on an object — the model finds its outline. The pink tint shows your selection.
  3. Lift it to a layer. Press "Lift selection (cut)". The object becomes its own layer; "Clean edges on lift" automatically removes color fringing at the cut.
  4. Export. Hit "Layered .psd" in the right panel. It opens in Photoshop, Photopea, Affinity, GIMP, and Krita with named layers, groups, and a real layer mask on every lift.

Separate a subject from its background

One button instead of clicking — best for portraits, products, and pets.

  1. Pick your engine. "Separate subject" (free) runs in your browser — first use downloads the model, ~40 MB. "Separate subject (server · 1 cr)" uses our strongest model: noticeably better hair, fur, and fine edges.
  2. Press it. You get a Subject layer on top and the background underneath — nothing else to do.
  3. Complete the background. The background now has a subject-shaped hole. Press "Fill behind last lift (server · 2 cr)" and AI fills it, so both layers are independently usable.

Decompose a whole scene at once

Turn every object in the image into its own named layer.

  1. Auto-detect objects (server · 1 cr). AI finds every object, outlines each one, and lifts them all into named layers ("person", "chair", "dog"…) in a single pass.
  2. Or Split by color (free). No AI — slices the image into layers by its dominant flat colors. This is the right tool for logos, posters, and UI screenshots. Tune the region count with the slider.
  3. Reorder and rename. Drag order with the ↑ ↓ buttons, double-click a layer name to rename, ⤓ merges down, ● toggles visibility.

Turn flat text into editable text layers

Posters, memes, screenshots — recover the text as text.

  1. Detect text (free). OCR runs in your browser and boxes every line it finds.
  2. Lift the lines. "Lift N lines + erase (server · 2 cr)" creates editable PSD text layers AND removes the original pixels in one inpaint pass. "Lift text only (free)" keeps the original pixels.
  3. Edit in your tool. The exported PSD carries them as real text layers — change the words in Photoshop or Photopea.

Fix a selection by hand

When AI gets 95% of the way there.

  1. Add with the Brush (B). Turn on "Add to selection" and paint the missed areas.
  2. Remove with Erase (E). Paint over anything that should not be selected.
  3. Wand for flat areas (W). Selects connected same-color pixels. Tolerance controls how loose the color match is; "Contiguous" off selects that color everywhere.
  4. Invert when it is easier backwards. Selecting a plain background then pressing Invert is often faster than selecting the subject.

Batch: 10–200 images in one job

Pro and Studio. E-commerce catalogs, sprite sets, photo dumps.

  1. Open Batch from the header. Drop up to 25 images (Pro) or 200 (Studio).
  2. Each image runs the pipeline. Subject separation per image, 1 credit each — debited up front, refunded per item on failure.
  3. Download the ZIP. One archive with every cutout, named after your files. Files are automatically deleted from our servers within 24 hours.

Fastest way to learn: try it.

Open the editor