Start free. Scale without the token bill.
You pay in credits for work done, not per seat. One credit is one step on the small model. Bigger models cost more, and anything your own fine-tune answers is free.
- ✓200 credits in your first week
- ✓Voice and sandboxes for 7 days
- ✓Public Stage rooms, 1 series
- ✓1 seat, 1 organisation
- ✓Every model, and they speak
- ✓Unlimited series, 10 documents
- ✓Credits roll over a month
- ✓Our branding removed
- ✓Review gates and assigned tasks
- ✓5 seats, then $25 each
- ✓Bring your own model key
- ✓One fine-tune, then those runs are free
- ✓White-label Stage
- ✓15 seats, 10 client workspaces
- ✓Credits pooled across clients
- ✓Three fine-tunes, 5 sites
Unlimited seats and organisations. SSO, full audit log with export, a private solution catalogue, an on-premises option, and a person who knows your name.
What a credit actually buys.
One credit is one step on the small model. Everything else is a whole multiple of it.
Waiting for an approval is free. So is a scheduled check that finds nothing to do. You are charged for work, not for readiness.
The awkward ones.
Yes. Fifty credits a month, forever, no card. Your first week comes with two hundred, plus voice and Python sandboxes so you can see the whole thing before deciding.
Because charging per seat punishes you for letting more of your team use it. Credits track the work actually done. Seats are included, and extra ones are $25.
Nothing breaks. Runs pause and you can top up or wait for the next refill. Unused credits roll over a month on Starter and Pro, three on Agency.
From Pro upward, yes — OpenAI, Anthropic, Google, Groq, or a private model on your own infrastructure. The tokens are then your spend, so we only charge for the orchestration around them.
A model trained on your own work, running on hardware we already own. It answers the routine turns for nothing, so a mature account costs less to run at month twelve than it did at month one. Pro includes one.
No.