Plans and quota behavior
What each plan includes and how monthly credits work.
Every generation runs real analysis and real model work, so all plans meter generations per month.
| Plan | Price | Generations / month | Input modes |
|---|---|---|---|
| Free | $0 | 3 | URL only |
| Pro | $9/mo | 25 | URL, screenshot, URL + screenshot |
| Max | $19/mo | 60 | URL, screenshot, URL + screenshot |
How credits are counted
- A successful generation consumes exactly one credit, whichever input mode it used.
- A request rejected before work starts - invalid input, plan restrictions, a blocked URL - consumes nothing.
- If a run fails before any paid analysis begins, the credit is refunded automatically.
- Retries and duplicate submissions are deduplicated server-side: they can't double-charge you or create duplicate Library entries.
Resets
Quotas reset at the start of each calendar month (00:00 UTC on the 1st). Your account menu shows the exact reset time in your local timezone.
What's gated by plan
- Screenshot input (alone or with a URL): Pro and Max
- Do's and Don'ts + Responsive Behavior sections: Pro and Max
- The full design-system report (sections 1–8) and the AI Context summary are included on every plan.