Pricing
Simple pricing. Clear ROI.
Teams spending $50K+/mo on LLM APIs typically save 3-10x the subscription cost in their first month.
Paid Pilot
- ✓ VPC deployment assistance. We help you get SpendProxy running in your infrastructure
- ✓ Full cost audit of your current LLM spend with accurate billing calculations
- ✓ All 5 optimization engines activated and configured for your workload
- ✓ Architecture review and optimization recommendations
- ✓ Direct Slack access to the founder throughout the pilot
- ✓ End-of-pilot report: actual savings, optimization impact, go-forward plan
Ongoing
- ✓ Continuous cost tracking and optimization across all providers
- ✓ All 5 autopilot engines with full configuration control
- ✓ Live dashboard with cost breakdowns, attribution, and request logs
- ✓ Priority support and quarterly optimization reviews
- ✓ All updates and new engine releases included
SpendProxy is for you if
- ✓ You spend $10K+/mo on LLM APIs (OpenAI, Anthropic, Google)
- ✓ You don't trust the numbers in your current cost dashboard
- ✓ You need to attribute AI costs to product features or teams
- ✓ You want automated optimization without manual effort
- ✓ Data privacy matters and you need a VPC-deployed solution
SpendProxy is not for you if
- × You spend less than $5K/mo on LLM APIs. You don't need this yet
- × You need a multi-provider routing gateway. Use LiteLLM
- × You need prompt evaluation and testing. Use Braintrust
- × You want a free, self-serve tool. SpendProxy is commercial and hands-on
Common questions
What happens at the end of the pilot?
You get a full report: actual savings found, optimization engine impact, cost attribution breakdown, and a go-forward recommendation. If SpendProxy isn't saving you at least 3x the subscription cost, we'll tell you. You can walk away.
How long does deployment take?
The Docker container takes 5 minutes to deploy. Pointing your AI SDKs at SpendProxy is a one-line environment variable change. Most teams are fully running within a day, including the initial cost audit.
What if I have custom models or providers?
SpendProxy supports OpenAI, Anthropic, and Google out of the box. Custom model pricing and additional providers can be configured during the pilot. If you have a specific need, bring it up on the first call.
Is there a long-term contract?
No. The pilot is a one-time $2,500 engagement. The ongoing subscription is month-to-month at $1,500/mo. Cancel anytime.
What data does SpendProxy access?
SpendProxy sees the API requests passing through it, the same data your AI provider sees. It runs entirely in your infrastructure. No data leaves your VPC. No telemetry, no analytics, no phone-home. API keys pass through untouched and are never stored.
Ready to see accurate numbers?
Book a 30-minute call to discuss your current LLM spend and whether a pilot makes sense.
Talk to a Founder