hey, so this is roast, I didn't invent the rules. Overall it's not a bad landing, clear what it does and what's the value.

Part on the screenshot looks like a straw man comparison. On the right it says manual preview but it's clearly a different neural network-made preview. Also on the "good" example head almost touches the edge which is maybe eye-catching but not in a way that wants me to use it.

Next slide says that if you make preview manually you save hours of work per each, and shows infinity symbol. I don't believe it's true, especially the infinity part.

Everything is put on a card. What doesn't help is the body is wrapper into another container, so it looks like two levels of hierarchy without a reason. Sometimes you add third level with a different background. That's probably things you actually wanted to make cards.

All headings have the same level.

Only one of the verified reviews goes to trustpilot. What verifies the others?

I don't believe trial plan with smaller models will make a good impression of the product but maybe I'm wrong.

Cheers and good luck with your app, makes total sense to not make previews manually.

The screenshot text and placements throw me. You mention "Perfect text placement" but it looks off to me. Might be the overlap between the "p" and the "W" but it doesn't sell me on the overall quality.

You might consider swapping the images so they flow from left to right which is amore natural "before (other products)" to "after (our product)" with possibly an arrow in between them or a visual clue, so people understand immediately understand it.

Good luck!