Most AI services don't tell you what you're using. You get a monthly bill. Maybe a vague "credits remaining" counter. If you're lucky, a bar chart that updates once a day.
We think that's backwards.
What You See in Your Stomme AI Dashboard
Every Stomme AI customer gets a real-time usage dashboard. It shows exactly how much of your monthly allowance you've used — broken down by provider:
- Claude tokens — the primary AI powering your agent
- ChatGPT tokens — used for specific tasks where a second model adds value
- Web searches — when your agent goes online to research something
You see the numbers. You see the percentage. You see when your usage resets. No surprises at the end of the month.
Cached Tokens Don't Count
Here's a detail that matters: when your agent reuses previous conversation context — which happens constantly in normal operation — those tokens are cached by the AI provider. Cached tokens cost a fraction of fresh tokens.
We exclude them from your usage count entirely.
This isn't a minor technical footnote. Cached tokens can represent 30–60% of your agent's total token throughput. By excluding them, your effective allowance is significantly larger than the number on the tin.
We could count them and make our margins look better. We don't, because that would be dishonest.
Why This Matters
AI pricing is confusing by design. "Unlimited" plans have fair-use policies buried in footnotes. Per-seat pricing hides per-token costs. Credit systems obscure what a "credit" actually buys.
We chose a different approach:
- Fixed monthly price. You know what you're paying before the month starts.
- Clear capacity. Your plan includes specific token and search allowances — listed on the pricing page, visible in your dashboard.
- Soft degradation. If you hit your limit, your agent doesn't stop. It switches to lighter models and keeps working. You're never cut off.
- Full transparency. You can check your usage any time. Real-time. Broken down by provider.
This isn't complicated. It's just honest.
What Happens When You Hit Your Limit
Your agent keeps working. It may use lighter models for some tasks, which means responses could be slightly slower. But it doesn't stop, it doesn't lose your data, and it doesn't charge you more.
If you consistently hit your limit, we'll suggest upgrading — not with a popup you can't close, but with a clear note in your dashboard. You decide. No pressure.
And if you don't want to upgrade? Your usage resets at the start of your next billing period. Back to full speed. No penalty for staying on your current plan.
The Principle
We sell a service, not a mystery. You should know exactly what you're getting, exactly what you're using, and exactly what happens when you reach the boundary.
That's not a radical idea. It's just not common in AI.
Stomme AI runs locally on your hardware. Your conversations and files stay on your machine. Your usage dashboard is part of your account at stomme.ai — the only part of the service that lives in the cloud.