Every artist has a gap.

For twenty years, closing that gap meant learning software. Then more software. Then deeper software. Tools grew a thousand features and a hundred panels, and somewhere along the way the skill being tested stopped being imagination and started being navigation.

You know the feeling. You picture something in ten seconds. Then you spend three days fighting a property inspector to get halfway there, and you ship the halfway version, because the show is on Friday.

That isn't a skill problem. It's a tool problem.

Power stopped meaning what it used to

Somewhere the industry decided that a more powerful tool meant a tool with more in it. More nodes, more parameters, more ways to do the same thing. So we got software where every single thing is possible and almost nothing is easy — where the answer to "how do I make this look like that" is a forty-minute tutorial and eleven settings you'll never touch again.

That trade made sense when computers were the bottleneck. It doesn't any more. The bottleneck now is the distance between having an idea and seeing it on a screen.

AI closed the gap, then got it backwards

Generative tools arrived promising exactly this: describe it, get it. And for a moment it looked like the answer.

But most of them closed the gap by taking the pen. You type, it decides. If what comes back isn't what you meant, your only move is to type again and hope. Roll the dice until something lands, then convince yourself that's what you wanted.

An artist doesn't want a result. An artist wants a result they chose.

A finished, unopenable image is not a creative tool. It's a vending machine. The moment you can't change the thing you were given, you've stopped being the author of it.

So we built the other half

In xRLive, the AI doesn't hand you a finished render. It hands you a real visual — actual geometry, actual materials, actual controls — and then gets out of the way. You move the camera. You change the light. You pull a slider and watch it respond, live, at show framerates.

And only the sliders that matter. Every visual exposes the handful of parameters worth touching, because somebody decided which ones those were. That's the part most software skips: choosing. A thousand exposed properties isn't generosity, it's an unmade decision handed to you at the worst possible moment.

The AI is there to get you past the blank page. Everything after the blank page is still yours.

What we believe

• The idea is the hard part. Everything else should get out of the way.
• Speed is a creative feature. When an iteration costs three days you make the safe choice. When it costs three minutes you take the risk — and the risk is usually where the good work is.
• Fewer controls, chosen well, beat a thousand controls. Curation is the work. Passing it to the user isn't power, it's abdication.
• You should be able to change anything the machine made. If you can't open it, you don't own it.
• Your work is yours. Your projects and your renders live on your own machine. We don't host them, we don't train on them, and we can't see them.

Who this is for

People who make things for a room full of other people — motion designers, video programmers, visual artists, the person running the screen behind the band. People with more ideas than time, who are tired of the tool being the reason the idea got smaller.

If you've ever finished a job thinking it was better in my head, we built this for you.