Pricing

Pay for tokens. Not for seats.

Every call is itemised to the cent — per model, per token, in and out. Start prepaid pay-as-you-go, or pick a plan with monthly included credit when your volume grows. Figures are illustrative; final plans may change.

Or pick a plan

Monthly included consensus credit, saved key-sets and higher limits — on top of pay-as-you-go, never instead of it.

Pay-as-you-go

€0

No subscription

Prepaid balance. You pay only for the tokens you spend, plus a flat consensus margin.

  • →No monthly fee, no minimum
  • →Consensus, diff, best_of and raw modes
  • →Per-model token, cost and latency on every call
  • →BYOK: provider rate + €0.01 per call
Start pay-as-you-go
Most popular

Starter

€19/ month

Recurring monthly · cancel anytime

For steady BYOK use. A monthly plan with €19 of included consensus credit and key management.

  • →€19 included consensus credit each month
  • →Saved BYOK key-sets
  • →Usage dashboard and budget alerts
  • →Pay-as-you-go beyond the included credit
Start this plan

Scale

€79/ month

Recurring monthly · cancel anytime

For teams shipping consensus in production. More included credit and higher limits.

  • →€79 included consensus credit each month
  • →Higher rate limits and more saved key-sets
  • →Priority support
  • →Pay-as-you-go beyond the included credit
Start this plan

Illustrative — final plans may change.

Pricing

You pay for tokens. Not per seat, not per call, not a flat tax.

A 3-model consensus on a typical question is ~€0.01–0.04. Every cent is itemised — per model, per token, in or out. Nothing hidden.

Single model (managed)
Typical cost / call: ~€0.003
What you can see: exact in/out tokens × that model's rate
3-model consensus (managed)
Typical cost / call: ~€0.015–0.04
What you can see: a line per proposer + the judge
Bring your own key (BYOK)
Typical cost / call: provider's exact rate + €0.01 flat
What you can see: you pay the provider direct; we add one cent

Accurate to the last token. Fully traceable.

Every cent, itemised

illustrative example
model                 in      out     cost
──────────────────────────────────────────────────
claude-haiku-4.5      812     540     €0.0041
gpt-4o                812     610     €0.0072
gemini-2.5-flash      812     498     €0.0029
judge (gpt-4o)        —       240     €0.0038
──────────────────────────────────────────────────
orchestration                         included
total                                 €0.0180

Accurate to the last token · your real receipt contains your exact counts

Estimate your cost

3
110
1
15
500
10050000

€0.0173

per call

€8.65

per month

Estimates based on representative tier-B model rates · actual costs depend on your model mix and prompt length

What counts as a call?

A consensus call is one prompt fanned out to your chosen models plus the judge. You are billed for the tokens each model and the judge actually use — never a flat per-seat fee. Included plan credit is spent first; anything beyond it falls back to pay-as-you-go.