Wait times — what to expect

Last updated: 2026-05-10

Printytron is honest about this: it's not always fast. Here's roughly what happens on each prompt and how long it takes.

A simple part from a simple prompt (“a 30mm cube with a hole through it”) — usually 8–20 seconds end-to-end. The AI builds the model, your browser renders it.

A detailed one-shot from a long, specific prompt — 30–90 seconds is normal. Could be longer. The AI is doing real reasoning, then a vision check on the result, sometimes a silent revision before showing you. This is the trade — you typed a paragraph, you skip three rounds of refinement.

A small refinement on an existing model (“make it 5mm thicker”) — usually 10–25 seconds. Faster because the AI already has the context.

When something goes wrong silently — if the AI's first attempt has a problem, it spots the issue and tries again. You see a “Building model…” overlay the whole time. You won't see the error. A couple of quiet retries can add another 10–30 seconds before it shows you what it could.

The vision check — after every model, Printytron captures three angles and sends them to a separate AI that grades whether the result matches what you asked for. If it doesn't, the model gets revised silently. You only see the final answer. This costs a couple of seconds per attempt.

The trade-off

Slower than typing. Faster than CAD.

What you'll see during the wait: an animated “Building model…” overlay on the viewer. Sometimes the chat shows the AI thinking. Don't refresh — refreshing keeps your work but cancels the in-flight build. Just wait it out.

If something takes longer than ~3 minutes, something has gone wrong. Refresh, try again, and consider whether your prompt could be shorter or more direct.


← Back to the guide