Free
Prove the concept
$0/mo
- 1 application
- Fair-use API rate limits
- 1 MCP endpoint
- Basic Glass Box tracing
- 1 member
- Community support
Bring your own AI.
Billed per app.
Your client or your API key — you keep your LLM spend; we host the app. Billed per app, not per usage.
Per application
Fair-use API rate limits
No usage metering
01 — CONNECT A CLIENT
Point your AI client at OBTO
Connect Claude Desktop, Cursor, ChatGPT, or any MCP client to your OBTO endpoint. Your client talks to its own model — OBTO never touches your inference.
02 — BRING AN API KEY
Or bring your own key
Prefer to run it headless? Point OBTO at an OpenRouter, Anthropic, or OpenAI endpoint with your own key. Same deal — your spend stays with your provider.
Either way, you keep your LLM spend; we host the app. OBTO bills per deployed application, not per usage.
Team
Scale your agents
$49/mo
- 5 applications+$8 / additional app / mo
- Higher API rate limits
- Standard Glass Box tracing
- SSO, roles & audit trails
- Up to 3 team members
- Email support
Business
Production at scale
$149/mo
- 20 applications+$6 / additional app / mo
- Unlimited API calls (fair use)
- Advanced Glass Box tracing
- SSO, roles & audit trails
- Up to 15 team members
- Priority support + SLA
Enterprise
Total architectural control
Custom
- Unlimited applications
- Self-hosted on your K8s
- Bring your own models
- Custom MCP integrations
- Dedicated support + SLA
Prices are per workspace, per month. Need the fine print on overages and fair-use limits? See full pricing details →
What's an application?
A deployed app is its own unit: its own MCP surface, routes, data, and Glass Box audit trail. Your plan includes a set number of apps — add more anytime at your plan's per-app rate. That's the only dimension we bill on.
Straight answers
The short version
The essentials most people ask before starting. Full breakdown on the details page.
Do you charge for AI usage?
No — never. You bring your own AI, so your model spend stays with your provider. OBTO only bills per deployed app.
Which providers can I use?
Connect a client (Claude Desktop, Cursor, ChatGPT, or any MCP client), or bring a key for OpenRouter, Anthropic, or OpenAI.
What happens if I need more apps?
Add them at your plan's per-app rate — +$8/app on Team, +$6/app on Business. Enterprise is unlimited.
Can I self-host?
Yes, on Enterprise — the full OBTO runtime on your own Kubernetes, with your own models. See the details →
Start free. Bring your own AI.
One app, fair-use API rate limits, and Glass Box tracing — no credit card required.