PRICING · BRING YOUR OWN MODEL

Priced in loops.

One loop per check. You pay for a verdict on your output, not for retries, and never for tokens: you bring your own model and we never proxy it. The unit and the calculator are public, so the bill is reproducible.

The numbers

100 free loops every month.

No card to start. When you run out, a pack is $20 for 10,000 loops ($0.002 each), and the card is only asked for at purchase. Sign in with GitHub to get a key and a balance.

Get a key → Try it free first

The rate ladder

Three tiers, and they run the opposite way to cached tokens: you pay less when you give back, more when you keep things private.

lowest rate

Pool

credit for contributing

Let the abstracted rules doloop learns from your failures flow back to the shared library, and your rate drops. You pool the lesson, never the data. (Pooled and looped, fittingly, are the same six letters.)

live today

Plain

$0.002 per loop

The base rate. One loop per check, deterministic verdict, reproducible. 100 free a month, then $20 per 10,000.

premium

Private

your own ratchet

Your house rules accumulate in a tenant-scoped memory that is yours alone, never shared, owned and exportable. (Live today for recurring extraction; expanding to the other donkeys.)

Why a loop, and why you can check the bill

A loop is one judged atom: one check on one piece of output. Not a token (those are your model's, and your cost, not ours), and not a retry (you are not billed for the donkey handing work back). Because we never proxy your tokens, there is no margin for us to take on your model spend.

The unit is defined in public code. The loop calculator is open, so anyone can reproduce a bill: loops times the published rate, both checkable.

The loop calculator (loops.py) → The public repo →