Simple, transparent pricing
Pay only for what you use. Attestation included on every request.
Free
100K tokens/mo
10 req/min
- Attestation receipts on every response
- OpenAI-compatible API
- Community support
- 1 API key
Pro
Unlimited tokens
60 req/min
- Everything in Free
- GPU inference (NVIDIA A10)
- Compliance evidence export (PDF + JSON)
- Email support
- Unlimited API keys
- Usage analytics
Enterprise
Unlimited tokens
Custom rate limits
- Everything in Pro
- VPC peering / private endpoint
- Bring your own key (BYOK)
- Custom MAA attestation policy
- Dedicated support engineer
- Custom SLA
Billing FAQ
How are tokens counted?
Tokens are counted using tiktoken (cl100k_base encoding) on both input (prompt) and output (completion) tokens. This matches OpenAI's counting methodology.
What happens when I exceed my monthly limit?
Requests will return a 429 error with a clear message. You can upgrade your plan at any time — the new limit takes effect immediately.
Can I upgrade or downgrade mid-cycle?
Yes. Upgrades take effect immediately. Downgrades take effect at the start of the next billing cycle so you keep your higher limit for the current month.
Do unused tokens roll over?
No. Token allowances reset at the start of each billing cycle.
What's the difference between monthly and annual Pro?
Both plans include identical features. Annual billing costs $200/year ($16.67/mo effective) vs $20/month — saving you $40 per year.
Need more? Let's talk.
VPC peering, custom attestation policies, BYOK, and dedicated support.
Contact sales