Run Promptflint at your scale, on your terms.
Self-hosted by default. SSO, audit logs, dedicated capacity, custom endpoints, and a phone-support SLA engineered around the way your team actually operates.
Pricing
Send a note below and a founder routes the conversation — within one business day, often sooner.
What ships at the Scale tier
Five things the public tiers don't include.
Every item below comes with the deployment. None of them require you to debug a CLI flag at midnight.
Self-hosted deployment
Run the proxy inside your VPC against your own provider keys. We deploy, you operate, your data never leaves the perimeter.
SSO + immutable audit logs
Wire the dashboard into Okta, Entra, or Google. Every rule, kill-switch, and token-savings event is signed and queryable for compliance.
Dedicated capacity + SLA
Reserved inference lanes, hot caches, and an uptime commitment negotiated against your traffic shape — not a percent off a shared tier.
Custom provider endpoints
Bring a private OpenAI-compatible host, an Azure deployment, or a self-hosted model. We wire the OpenAI/Anthropic contract to whatever you operate.
1-hour phone support
A phone number, a named engineer, and a one-hour response window. Production incidents do not wait on ticket queues.
Get in touch
Tell us about your team. We'll route the conversation.
Volume, providers, residency, and procurement requirements — the more you send, the less back-and-forth we need. We reply within one business day.
- Drop a note — the form on the right reaches a founder directly.
- Schedule a call — for anything procurement, residency, or integration-shaping.
- Get a tailored plan — volume, terms, and an SLA engineered for your shape of traffic.