A page costs what it took to read.
Pick by what you crawl, not by a feature grid.
- Free$01,000 credits, once
- Starter$29100,000 credits a month
- Growthmost teams$99500,000 credits a month
- Scale$3492,500,000 credits a month
- Enterprisefrom $1,500no ceiling credits a month
Annual billing: twelve months for the price of ten. Credits arrive monthly on paid plans; the free 1,000 are a one-time allowance and do not renew. Every plan includes change detection, sitemaps, destinations, exports, webhooks, the SDKs and the MCP server.
Six modes, and the ladder picks the cheapest that works.
A refused page climbs one rung, never two. A page that needs more than your tier allows is not charged and is marked TIER_LIMIT — “this page costs more than you allow” — with the rung that would have read it. maxCredits on a request caps any single page.
- 1Fetchhttp1
- 2Fetch with sessionhttp2
- 3Renderbrowser4
- 4Render, residentialstealth16
- 5Render, solvedstealth22
- 6Unlockedstealth40
Render, residential includes 1 MB; each started megabyte past the first is 4 credits.
Estimate it before you sign up.
Four sliders and the plan that covers them. The weights are the schedule above; the answer is an estimate, and the ledger on every run is the bill.
- 400
- 1,600
- 202
- 16 / 24
- 500
- 300
- 4
- 0
- ~120
- A page costs what it took to read.
- You pay for pages, not for walls.
- A site you have read before starts where it last answered.
- Change detection, sitemaps and destinations are included.
- Extraction runs on your own model key.
Search and the agent, priced the same way.
A search pays for the results page it read; an agent pays for the pages it read and its model’s tokens. Refused pages are free here too, and every agent run stops at a cap you set. The web search API · the AI research agent · rank tracking
- 4
- 1
- 0
- 0
- per page
- per page
- price + 20%
- 1 / 1k tokens
- 0
- 2,000
- per row
- 0
- as one run
- ~120
The ceiling you set, per project.
A tier is the most a page of a project may cost. The plan caps the tier; the project may ask for less than its plan allows, never more.
Per page, on top. And the things that never cost anything.
- +1
- min 2
- 1 / 1k tokens
- 1 per sitemap file read
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
What people ask before they put in a card.
What is a credit?
The unit of everything. A plain fetch is one credit, a fetch with a site’s session two, a real Chrome four, a residential address sixteen (with a megabyte included, then four more per started megabyte), a solved widget twenty-two, a rented unblocker forty. The response to every request says the mode each page was read in and what it cost.
What costs nothing?
A page the site refused, whatever the ladder spent finding out. A 404 or 410. A page served from your cache. A copy carried forward when the sitemap says the page has not changed. Change detection, sitemaps, link-graph reports, destinations, exports and webhooks, on every plan.
Can I cap what a page or a project may cost?
Three ways. The plan caps the tier; a project may ask for a lower tier than its plan allows; and maxCredits on any request caps one page. A page that needs more than you allow is not charged and comes back TIER_LIMIT with the rung that would have read it. What your workspace has left caps it too: a page only a dearer engine could read comes back BUDGET_EXHAUSTED, also at zero.
What happens when the credits run out?
Runs stop and say so; nothing is read on credit. The free 1,000 are a one-time allowance and do not renew — for more, move to a paid plan. On a paid plan there are two ways past the allowance: buy credits ($5 a pack — 15,000 on Starter and Growth, 25,000 on Scale — drawn after the allowance and never expiring), and on Growth and Scale automatic overage at $0.30 or $0.20 per thousand, billed with the next renewal, up to a cap you set on Billing (twice the allowance by default). Scale also carries a month’s unused credits into the next.
Is there an annual price?
Yes — twelve months for the price of ten. Credits still arrive monthly.
How do I pay, and in what currency?
Through Razorpay, in USD or, for a customer in India, in INR — Starter ₹2,499, Growth ₹8,499, Scale ₹29,999 a month, plus 18% GST added at checkout. Cards, and for INR also UPI and net banking. MeshArc never sees your card; Checkout does. Every payment gets a numbered tax invoice on Billing — CGST and SGST for a customer in Madhya Pradesh, IGST for the rest of India, zero-rated as an export for a customer abroad — with your GSTIN on it when you give it under Billing. The supplier is Discover Webtech Private Limited, Indore.
Can I change or cancel?
Any time, from Billing. An upgrade is applied at once — the difference for the rest of the cycle is charged and the month’s credits adjusted the same way. A downgrade takes effect when the cycle ends and can be called off until then. Cancelling ends the plan at the cycle end; the workspace goes to the free plan and keeps its projects and data.
Is the model bill included?
No, by design. Structured extraction and embeddings run on your own model key; MeshArc charges one credit per thousand tokens for the pipeline and you pay your provider for the tokens. Nothing is marked up.
What does a web search cost?
A results page costs what it took to read: typically 4 credits for Google’s, read in a real browser, and 1 when Bing answers instead. A results page every engine refused is free, and the same search again within the hour (ten minutes for news) is answered from the last one at no charge. Ask for the results’ pages too and each is charged like any page.
What does an AI agent run cost?
The pages it reads, at what each took to read (a refused page is free), and the model’s tokens at the model provider’s price plus 20%. On your own model connection the tokens are billed to your key and MeshArc charges 1 credit per 1,000. Every run has a cap — 2,000 credits unless you set another — and stops there with what it found.
What does running the agent over a table cost?
Each row is one agent run, paying only for what it used up to the cap you set per row, and the table never spends more than rows times that cap. Checking the table first is free and says about what it will cost. A table holds up to 25 rows on the free plan, 200 on Starter, 500 on Growth and 1,000 on Scale.
How do I keep scheduled searches and agents from overspending?
Each monitor can have a monthly cap, and an agent monitor a cap per run. Before you save, the estimate says about what a month will cost; once a month’s runs reach the cap, the next is skipped until the month turns. A workspace with nothing left skips its runs rather than running on credit.
Start on the free 1,000 and read the ledger.
Browser tier included, no card. Every page in the response says what it cost and why.