Free until you grow out of it.

A search is a request to the ranked-results endpoint. A chat is a request to the AI answer endpoint. They're metered separately — Pro and Enterprise plans include a monthly allowance of each.

One page processed by a crawl or a publish. Indexing a 5,000-page site once uses 5,000 page updates; re-crawling it uses 5,000 more, so what you consume depends on how often your content changes and how often you crawl — not just how big your site is. Pages we skip before processing (redirects, error pages that never reach the index) don't count.

Yes. Upgrades take effect immediately; downgrades at the end of the billing period. Stripe handles the math.

Indexing stops and AI answers pause, and search keeps working for your visitors. To keep going past the allowance, turn on overage in billing settings: page updates beyond the plan bill at US$2.00 per 1,000 and AI answers at US$8.00 per 1,000. Searches are never billed. We email you at 80%, at 90% and when an allowance is used, so it is never a surprise.

Yes, and we require it. Turning on overage means setting one monthly cap that covers page updates and AI answers together. When the cap is reached we stop crawling and pause AI answers rather than keep charging, and we email you at that point. Overage is off until you turn it on, so nothing is ever billed beyond your plan price unless you asked for it.

Sustained volumes well beyond Enterprise — or more than 25 sites — are handled on a custom contract with an allowance sized to your content. Talk to sales.

Out of the box, QuantSearch is fully managed. For specific cases (existing AWS Organization, data sovereignty rules, air-gapped environments), we scope a deployment into your own AWS account on a case-by-case basis. Talk to sales.

Compute and the search index run in AWS regions selected for low-latency delivery. AI inference runs on AWS — prompts and responses aren’t retained or used to train models. We can pin a region for your tenant on request.