Pricing

Two plans. The safeguards are in both.

Plans are priced on model spend, because that is what actually varies — a text run costs a fraction of a cent and an image run costs many times that. The enforced cap, the kill-switch and the full decision record are not upsells; they are on the free tier too. Paying buys room to run, not permission to be safe.

 
Free
$0
Enough to point it at a workflow that already costs you money and see what the trace says.
  • $1 of model spend a month
  • Fair use: 10 runs
  • Enforced monthly spend cap
  • Capability kill-switch
  • Full per-step run trace with real cost
  • Routing across seven providers, with fallback
Start free — no card

USD, billed monthly, cancel any time. No team plan yet — Prompt Tornado is single-account today. If you need seats with separate budgets per person or per client, tell me and it moves up the list.

What "enforced" means here

Most tools alert you after the spend. This blocks the call before it happens.

Before dispatch
The cap is checked at the chokepoint

Every model call — text, image, audio, video — goes through one dispatch path that reserves budget first. If the reservation fails, the call is never made.

Under concurrency
Two runs cannot both slip past

Reservations are atomic in the database. Simultaneous runs starting at the same instant cannot each read "budget remaining", each see room, and both proceed.

Priced before it runs
Media is quoted at what media costs

Image, audio and video steps are estimated per generation rather than as text, so the amount held against your cap matches what the run actually costs.

What you are not buying yet

Written down so you can decide with the real picture. If any of these is a dealbreaker, don't pay yet — tell me and I'll say when it lands.

One account, no seats

There is no multi-user mode, so no per-person or per-client budgets. The spend cap covers the whole account. This is the most-requested gap and the most likely next build.

No SSO, no SOC 2 report

Sign-in is email or social, and there is no audit report to hand a security team. The architecture is public and the run record is exportable, which is not the same thing.

Redaction detects, it does not yet mask

PII and secret detection runs on outbound prompts and records what it found. Masking before send exists behind a flag and is off in production.

Small and new

Built by one person, in public, with a public changelog you can use to judge the pace. Support is direct email and usually same-day — the upside of the same fact.

Frequently asked

What happens when I hit my spend cap?
The run is blocked before the model call is made, not after. The cap is enforced at the point of dispatch and holds under concurrent runs — two runs starting at once cannot both slip past it.
Are model costs included, or do I pay providers separately?
Included, up to your plan's allowance. Past that, runs are blocked by default — you can opt into overage at provider cost plus 10%. Every run shows its exact cost, so nothing has to be reconstructed from an invoice later.
Which limit will I actually hit — spend or runs?
Spend, almost always, which is why it is the headline number. Measured across every run this platform has executed, the average costs $0.0091 and the median $0.0030, so $10 is roughly a thousand ordinary text runs. Generate images or audio and you will reach $10 far sooner — the most expensive single run on record is $0.30. The run count is a fair-use ceiling for very cheap workloads, not the thing that stops most people.
What counts as a run?
One submitted workflow, however many steps and models it uses. A five-step run touching three providers is one run.
Do you have team or multi-seat plans?
Not yet — Prompt Tornado is single-account today, so budgets are per account rather than per person. If you need seats, say so; it is the clearest signal for what to build next and it moves up the list.
Can I cancel?
Any time, from billing settings. If it isn't useful in your first 30 days on a paid plan, email me and I'll refund it in full — no process.
Does the evidence export make me EU AI Act compliant?
No — and be wary of anyone selling that. It produces a timestamped, exportable record of inputs, model decisions, the policies active at each step, and human approvals. Whether that maps to an obligation you actually have is a question for your counsel, not a checkbox anyone can tick for you.

Ten runs, no card.

Point it at a workflow that already costs you money and read the trace.