levi×athan

Upscale an image as large as this device can hold — no fixed ceiling — then optionally raise every tile's detail through Google Gemini, using your own API key.

upscale runs on this device worker engine

Source

—

Resampling plan

catmull-rom

Method

Fast interpolation. Predictable, runs on anything, no download.

Target size

Quality

Filter

Catmull-Rom. Crisp edges, the safe default for photos and art.

Enlarge in steps of2.0×

Several small passes hold detail together better than one big jump. 1× does it in a single pass.

Unsharp mask35%

Radius0.8

Threshold2

Output

Gemini clarity pass

Sharpen every tile through Google Gemini after upscaling

The upscaled image is cut into ~1024px tiles and each one is sent straight from your browser to Google's Gemini API — using your own key — asking it to de-blur and restore detail without changing the content. Then everything is stitched back together here. This step leaves your device; the local upscale above does not. Tiles are independently re-imagined by the AI, so faint seams between tiles are possible, especially on busy or high-contrast images.

Your Gemini API key

Stored only in this browser (localStorage), sent only to Google, never to us. Get a free key at aistudio.google.com/apikey.

The upscale itself never leaves this device — there is no server, no account, no upload for that step. Resampling uses a separable Catmull-Rom / Mitchell / Lanczos-3 filter on premultiplied alpha, applied over several passes, with an optional AI reconstruction pass and Gaussian unsharp mask, all in a background worker. If you turn on the Gemini clarity pass, tiles of the result are sent directly from your browser to Google's Gemini API using your own key — that part does leave your device, by design, so the cost and the data stay between you and Google, not us.

More free tools from Massey Infinite LLC: see the full toolkit