Platform access is per person. Inference is separate, and you choose whether it comes from your included credits or from your own provider accounts.
For trying it on a real repository, not a toy one.
For a team running agents on production repositories every day.
Full isolation, governance and scaling.
Platform access is what you pay us per person. Inference is what the models cost. Route it to platform credits, or to your own provider accounts and subscriptions, and see the split in Usage.
Token spend on platform-billed models. Enough to take several real briefs through a full Standard workflow rather than one demo.
Yes. Connect your keys whenever you like, and your included credits stay unspent until you use platform-billed models.
Anyone in your organisation who can create runs or answer gates. Read-only reviewers do not need a seat.
Runs stop at the cap and tell you why. Nothing continues quietly past a limit you set.
No. Agents open pull requests and can wait on CI. Your team merges, on every plan.
Yes, on Enterprise: a sandboxed on-prem or VPC deployment. Talk to us about your constraints and we will be straight with you about fit.