Vrge Managed AI
Vrge itself is a one-time purchase, and its AI is free either way: run it on your own machine with Ollama, or plug in your own cloud key. Managed AI is an optional add-on for people who'd rather not set up either, and it requires a Vrge license. Pay a flat monthly fee and we run the AI for you: hard token quotas, zero overage bills, cancel anytime. Access runs through your current billing period, no renewal after that. The app works identically if you never subscribe.
Managed AI · optional monthly
Starter
One mailbox at a normal pace — about 15 new emails a day for the observer to read.
- 1.5M AI tokens every month
- Redact-by-default on every cloud call
- Cancel anytime. No renewal, no retention nags
- Falls back to BYO or Ollama at cap
Pro
One busy mailbox, or two quiet ones — about 40 new emails a day.
- 4M AI tokens every month
- Per-user fair-use allocation for teams
- Admin dashboard: top consumers, projected usage
- Mid-cycle upgrade with prorated quota credit
Power
Three or four busy mailboxes — about 120 new emails a day.
- 12M AI tokens every month
- Per-user fair-use allocation + admin controls
- Priority routing to the fastest available model
- Hard quota · no overage
- Redact-by-default
- Cancel anytime · no renewal
- Excludes VAT / sales tax
More than 120 emails a day? Email us for custom pricing — same flat fee, same hard cap, sized to your actual mail volume.
How it works
When you subscribe, the desktop app routes AI calls through a Cloudflare Worker we operate at ai.getvrge.com. The Worker holds the upstream provider keys, enforces your monthly quota in a managed database, and forwards the redacted request.
The proxy is the source of truth. It checks your quota + subscription status before calling any upstream model. If you're over your cap, the Worker refuses the call with HTTP 429 and your app falls back to BYO keys or Ollama for the rest of the cycle. There is no “approximate” enforcement; the quota is a number in the database.
Cheapest-capable routing. You see a flat token quota. Internally, the Worker picks the smallest model that can handle each job: a fast, cheap one for classification and routine extraction, a mid-tier one for larger extractions and summaries, and the most capable one only for complex drafting. That spread between retail price and wholesale model cost is where the tier's gross margin lives, and it's why your quota goes further than buying the top model outright.
What we log (metadata only)
What we do not log: prompt bodies, completion text, schema definitions, or any user-identifying payload fields. This is a schema-level guarantee. The usage log table has no column for content, so no code path can leak it even by accident. The privacy invariant is enforced in the proxy test suite: every row written to the log is checked to have no content column. If someone adds one, a test fails and the deploy is blocked.
The Managed AI proxy records one row per forwarded call with the following fields — and no others:
| Logged field | Example |
|---|---|
| Your Vrge license key | |
| Timestamp | |
| Upstream provider | Anthropic |
| Model selected | |
| Input + output token counts | |
| Task type | classify / extract / summarize / draft |
| Status | ok / rejected_quota / error |
| Redaction mode that applied |
For teams: fair-use inside the quota
The team concern with any shared AI quota is one heavy user burning the month's budget in three days. The Pro and Power tiers solve this with a per-user allocation layer.
- Per-user caps.
- Admins can set a soft token cap per member, so one heavy user can't quietly consume the whole org's month. Caps are whatever the admin sets.
- Admin override.
- The admin dashboard has a per-user slider. Boost your power user; throttle the intern.
- 80% warning.
- When a user approaches their allocation, they see a banner in their own app.
- Top-consumers view.
- The admin sees exactly which teammate and which feature has burned the most tokens this month. No guessing.
- Mid-cycle upgrade math.
- Upgrade Pro → Power on day 20 with 1.5M used and you get
8M − 1.5M = 6.5Mremaining for the final 10 days, prorated billing. No gaming, no accidental starvation.
The six guardrails we won't break
These are public commitments. If a future feature violates one, that feature has a bug.
BYO keys stays free forever.
Every feature in Vrge works with Ollama locally or with your own Anthropic/OpenAI/Google key. Managed AI is convenience, never a gate.
Hard quota, zero overage.
When you hit your monthly cap, the proxy refuses the call. No 'reasonable usage' language. No surprise bill, ever.
Redact-by-default through our proxy.
The client applies redaction before sending. The proxy verifies it was applied for non-manual sources and refuses the call otherwise.
No prompt/response logging.
Metadata only. Schema-level invariant: there's no content column to leak.
Cancel anytime, no dark patterns.
One click through the Lemon Squeezy customer portal. Access runs through the current billing period, no auto-renewal after cancel. No retention nags, pause-first flows, or win-back emails.
Live quota meter + cost preview.
Settings → AI shows real-time usage. Manual actions whose estimated cost exceeds 5% of remaining quota prompt you first.
Or skip the proxy entirely.
Legal, medical, airgapped and regulated teams do not need the proxy at all. Run a local model on your own machine, or use your own cloud provider key, and no request touchesai.getvrge.com. That path has zero dependency on us, and it is free.
Team server
Running the team server on your own hardware is a separate thing, and it is available today. See the self-hosting guide.
Questions?
Read the Managed AI FAQs, see the privacy policy for Managed AI, or email us.