Kitana: What the 14 Tools Do, and Where They Fall Short

We make this app, so treat this as what it is: our own account, not a review. We have written it because a plain description of what the tools do — including what they do badly — is more useful than marketing copy, and because a self-review dressed as an impartial one would be dishonest.

What is in it

Thirteen photo tools. AI avatar, AI headshot, style transfer, background replacement, object removal, photo upscaler, text to image, social content, baby generator, 90s yearbook, age transform, cartoon, photo restore.

Photo to Video, which animates a still into a short vertical clip.

Four of the photo tools — text to image, headshot, background and style transfer — also run in the browser at photomaker.lol/studio, with no install.

What it is good at

Background replacement. The highest-value single edit and the most reliable. It removes every distraction in one step and does not touch the subject. See the guide.

Headshots from ordinary photos. Evening out lighting and replacing a background on a photo you already have. Restrained edits, which is the right way to use it — the reasoning is in the LinkedIn headshot guide.

Restoration and upscaling of old photos. Damaged prints and small scans are the cases where these models earn their place, because there is real structure to work from.

Having everything in one place. The practical argument for a fourteen-tool app is not that you need fourteen tools; it is that you do not export to a second app between steps.

What it is bad at

The same things everything in this category is bad at, and we would rather say so:

• Hands. Unreliable in stills, worse in motion.
• Legible text. Anything the model generates or repairs will look like writing and read as nonsense.
• Fine repeating patterns. Brick, tile, fabric weave. Expect broken rhythm.
• Anything partly transparent. Glass, water, smoke, sheer fabric.
• Accuracy of any kind. These produce plausible images, never verified ones. That matters for object removal over text and for upscaling a blurred face.

Who should not use it

Anyone who needs an edit to be exact. Product colour, a logo, a document, anything a viewer relies on. A layered editor gives you control; this gives you prediction. The comparison is in AI vs Photoshop.

Anyone printing large. Check output resolution against what the print needs before building a workflow.

Anyone who wants to adjust a result afterwards. Generative edits are whole outputs, not layers. If it is 90 percent right, you regenerate rather than nudge.

The limits, stated plainly

Free: one image on the web without an account, then two photo generations a day with one every four hours. Pro: six a day with no waiting, plus Photo to Video at ten clips a month, four seconds, 768 by 1344. Pricing is on one page.

Photo to Video runs in the apps, not the browser studio. Video clips are four seconds, vertical, silent.

What happens to your photos

Used to produce your result and nothing else. GPS and other EXIF metadata stripped before upload. Deleting a result in the app removes it from storage. Email support from the address you want removed and your data is deleted. The privacy policy is the document that binds, not this paragraph — and the checklist for evaluating any app on this is in AI photo apps and privacy.

Why a fourteen-tool app instead of one good tool

A fair objection, and worth answering rather than dodging.

The case against bundling is real: an app doing one thing usually does it better, and a long feature list is often a substitute for depth. We would not argue otherwise in general.

The case for it here is specific. Most photo jobs are two or three operations, not one — remove the distraction, replace the background, crop for a platform. Each hand-off between apps costs a re-encode, a re-upload and a decision about where the file now lives, and those costs are larger in practice than the difference between a good tool and a very good one.

Where that argument fails is when you need depth in a single operation. If your work is one kind of edit done precisely and repeatedly, a specialist tool or a layered editor will serve you better, and no amount of breadth compensates.

We would rather you knew which of those you are before installing anything.

How to decide without trusting us

Run the ten-minute test on your own photo. Four of the tools run in the browser with no account, which makes it free to find out whether this works for your face and your photos — which is the only question that matters and the one we are least qualified to answer for you.