- Setup is one URL change. Point your existing SDK at Phantm. Streaming and tool calls work the same. You can be live in minutes.
- You only pay when we save you money. No platform fee, no seats. We charge 25% of the savings we can prove on your traffic. No savings, no bill.
- Every change is logged. Each optimization has a reason code and can be rolled back. Your engineers can audit everything.
How the perk works
- Come through Rho Perks or book a call below. We check eligibility during onboarding.
- Change one base URL. Your first week is free with no commitment.
- For 30 days you pay no Phantm fees. Every dollar we save you is yours.
- After that, our standard pricing applies: 25% of verified savings, or 18% with data sharing. Cancel anytime.
Eligibility: Active Rho customers.
FAQ
Is this really zero risk?
Yes. We charge a percentage of savings we can prove, request by request. During your perk window that percentage is zero. After it, if the savings stop, so do the fees.
Will it degrade output quality?
No. Optimizations are measurable and reversible. You see a diff and a reason code for every request, and you can roll back any policy instantly.
Does it break streaming or tool calls?
No. Phantm is a proxy that speaks the OpenAI API. Streaming and function calling pass through unchanged.
What data do you see?
Requests pass through the gateway. Content logging is off by default and retention is configurable. A security packet is available on request.