Skip to content
Get started

Pricing and Plans

How Tabstack billing works: what a credit buys, what each endpoint costs, the three plans, and how overages are handled.

Tabstack is credit-based. One currency covers the whole platform: every API call spends credits, and the cost depends on which endpoint you hit and how hard you ask it to work. There are no per-seat fees and no separate charge for extraction versus transformation versus research. You buy credits and spend them on whatever mix of endpoints your workload needs.

Every account starts with 10,000 free credits, no card required. Pricing is separate from rate limits: rate limits cap requests per minute, pricing is what those requests cost. The two never interact.

Credits are spent per action. An action is one unit of work the platform performs. The cost of each action is set by its endpoint. You can see each action’s credit rate next to the run button in the console before you run it. For Extract and Generate that rate is the full cost of the call; for Automate and Research the total depends on how many actions run (see the FAQ below).

EndpointWhat it doesCost per action
Extract (Markdown)Turn any URL into clean markdown10 credits
Extract (JSON)Turn any URL into schema-enforced JSON50 credits
GenerateTransform page content into a custom output100 credits
AutomateRun a browser-like task: click, scroll, fill, submit100 credits
Research (Fast)Autonomous multi-source answer with citations250 credits
Research (Balanced)Deeper research pass, more sources350 credits

How many actions a call spends depends on the endpoint. Extract and Generate are always a single action, so one call costs exactly the rate above. Automate and Research run a variable number of actions per call: each one chains as many actions as the task needs, and you pay for every action it runs. A Research call that touches more sources, or an Automate call that takes more steps, costs more than one that does less.

Credits are easier to reason about as work done. The Team allowance of 500,000 credits a month is roughly any one of:

  • 50,000 markdown extraction actions
  • 10,000 JSON extraction actions
  • 5,000 Generate actions
  • 5,000 Automate actions
  • 2,000 fast Research actions

Extract and Generate spend one action per call, so a call maps one to one onto those numbers. Automate and Research spend several actions per call, so a single call draws down more than one. Most real workloads are a mix, so treat credits as a shared budget across all endpoints, not a quota per feature.

Every plan includes access to every endpoint. What differs is how you pay for credits, which research modes you get, and your rate limits.

PlanPriceCredits includedOverage rateResearch modes
Individual$0 / monthPay as you go$0.35 per 1k creditsFast
Team$99 / month500,000$0.30 per 1k creditsFast, Balanced
Pro$499 / month3,000,000$0.25 per 1k creditsFast, Balanced

Individual is pay-as-you-go at $0.35 per 1,000 credits, with standard rate limits and fast research mode. Spend is capped at $99 a month, about 282,857 credits, so requests stop there rather than billing past it. For tinkerers and hobbyists connecting their systems to the web: start here, pay only for what you use, upgrade when your usage becomes predictable.

Team is $99 a month with 500,000 credits included, increased rate limits, and both fast and balanced research modes. The right plan once you are running real workloads and want a fixed monthly number instead of metered billing.

Pro is $499 a month with 3,000,000 credits included, the highest rate limits, and the lowest per-credit cost. For teams running production workloads at scale that want headroom.

Overages are on by default for Team and Pro. Going over your plan’s included credits does not cut you off and does not fail a request. You keep going, and the extra credits are billed automatically at your plan’s overage rate:

  • Team: $0.30 per 1,000 credits over your 500,000.
  • Pro: $0.25 per 1,000 credits over your 3,000,000.

Each plan’s overage rate is lower than the pay-as-you-go rate, so a plan is always cheaper per credit than Individual once you are using it. Whether your plan has an upper limit on overage credits is set per account; check the console billing page for your cap.

Trial and Individual accounts do not have overages. A Trial stops when its free credits run out. Individual stops at its $99 spend cap, about 282,857 credits. Neither bills beyond that point.

To opt out of automatic overage billing on Team or Pro, set a spend threshold from the console billing page, by percentage of your allowance or by a fixed dollar amount, and choose Notify and Stop to halt requests when you cross it. Notify Only alerts you but lets billing continue.

For custom credit quotas, dedicated support, and SLAs for high-volume teams, contact sales for custom pricing.

What is a credit? A credit is the single unit of usage across Tabstack. Credits are spent per action, and the cost of each action depends on the endpoint. Extract and Generate are one action per call; Automate and Research run as many actions as the task needs and bill for each. See the endpoint cost table for per-action costs.

What happens when I run out of credits? On a Trial, requests stop when your free credits run out. On Individual you are billed per 1,000 credits as you go up to a $99 monthly cap, about 282,857 credits, then requests stop. On Team and Pro, overages are on by default, so you continue at your plan’s overage rate unless you set Notify and Stop in the console. Any upper limit on overage credits for your plan is shown on the console billing page.

Can I set a spending limit? Yes. In the console billing page you can set a threshold by percentage or dollar amount, and choose to be notified or to stop requests when you reach it.

Which research mode should I use? Fast mode (250 credits) is the default and is right for most queries. Balanced mode (350 credits) does a deeper pass across more sources and is available on Team and Pro. See the Research guide.

How do I know what a call will cost before I run it? The console shows the per-action credit cost next to the run button on every endpoint, plus a running balance across the top of every page. For Extract and Generate that is the full cost of the call. For Automate and Research the total depends on how many actions the call runs, so the final cost is known once it finishes.

Plans, current spend, and credit balance live in the console, not in any API response. For request-rate ceilings rather than cost, see Rate Limits.