Vmake is one of those AI media tools that does fifty things on its website and one thing I care about: it removes watermarks and small artifacts from images and video, and it upscales both. It has a web app and a developer API, and I only ever use the API.

How I use it

Generated images come out of most models with a small mark in a corner. Before one goes anywhere it runs through a short Python script that calls Vmake's eraser endpoint, checks the result and saves a clean copy next to the original. The same script upscales small video clips, which is an async job: submit, poll, download. I run it on folders, not on single files, and I never open the website.

What surprised me is that the removal is clean on the kind of image I throw at it: product shots, portraits, plain backgrounds. The corner comes back with the texture the model would have put there, not a blur. Upscaling video is a slower story but the result is honest, no plastic skin. What annoys me is around the edges. Video jobs sit in a queue with no estimate, so a batch of twenty takes an unknown lunch break. The API pricing is not on the public pricing page; you learn it from the developer console. And the web app, if you do open it, wants to sell you face swaps, product videos and a dozen effects, which makes the one good tool hard to find. On a cluttered background the eraser still smears, and I finish those by hand.

Who it is for

People who produce or collect a lot of media and need a reliable clean-up step in a pipeline. If you have ten images a week, the free daily credits may be enough. Skip it if you only need a one-off fix, where any free web tool will do, or if you need pixel-perfect removal on complex scenes; that is still a job for Photoshop.

Related tools