Token-metered plans
One flat token budget per month across every model — prompt and completion counted equally. No surprise bills, no per-model math. Start free, upgrade when you outgrow it.
For trying Xevyo and light personal use.
$0/forever
No card required
Start freeIncludesFor regular users who want the full agent.
$19/month
or $190 billed yearly
Choose BasicEverything in Free, plusFor power users and demanding workloads.
$49/month
or $490 billed yearly
Choose PremiumEverything in Basic, plusPrices in USD. CAD supported at checkout. Tokens are counted flat — prompt + completion summed across all models. Image generation is metered separately.
| Feature | Free | Basic | Premium |
|---|---|---|---|
| Monthly token budget | 75,000 | 1,500,000 | 6,000,000 |
| Image generation | — | 50 / month | 300 / month |
| Model access | Fast path (SPARK) + standard agent | Full agent incl. research synthesis | Full agent + priority routing |
| Max tokens / request | 1,500 | 4,000 | 8,000 |
| Agent tool iterations / turn | 4 | 8 | 12 (max) |
| Conversation memory | Short | Standard | Extended |
| Document upload (RAG) | 3 active docs | 25 active docs | Unlimited (fair use) |
| Google integrations | ✓ | ✓ | ✓ |
| Support | Community | Priority email | |
| Annual discount | — | ~2 months free | ~2 months free |
Prompt and completion tokens are summed and counted equally across every model. No per-model pricing to decode — one number, one meter.
Paid plans reset on your Stripe billing anchor; Free resets on the calendar month (UTC). Your invoice and your quota window always line up.
When the budget runs out we stop inference and show an upgrade prompt — never a surprise overage bill. Predictable spend by design.
Image generation has its own monthly quota because it runs on diffusion GPU seconds, not tokens. Keeping the meters separate keeps both honest.
Agents, Agent Studio, Flow Studio, the API and private deployment are priced to how you use them. Tell us what you need and we will put together a plan.
The ready-made agents on the Platform: contracts, data rooms, social posts, email, documents, code review and calls.
Contact us
Industry agents for your own product, as an API we run or source code you run.
Contact us
Your own agents and workflows, and API access with a monthly limit on every key.
Contact us
The whole platform in your own environment, on-premises or air-gapped. Custom agents through Enterprise.
Contact us
Questions
A token is roughly ¾ of a word. We count both what you send (prompt) and what the model returns (completion), summed together — so a 1,000-token answer to a 500-token question uses 1,500 tokens from your budget.
Xevyo hard-blocks further inference and shows an upgrade prompt with your reset date. You are never charged an overage. You can upgrade instantly and your higher budget is live within seconds.
No. Budgets reset each month and do not carry forward. This keeps GPU capacity predictable and pricing simple. If you consistently need more, Premium or Enterprise is the better fit.
Yes. Upgrades take effect immediately with prorated billing handled by Stripe. Downgrades schedule at the end of your current period, and your quota adjusts at the next reset.
Billing runs on Stripe in USD (primary) with CAD supported at checkout. You can manage your card, invoices, and cancellation anytime through the Stripe Customer Portal.
No — Basic includes 50 image generations per month. Only the Free tier excludes image generation. Premium raises the image quota to 300 per month.
Agents, Agent Studio, Flow Studio, the API and private deployment are priced with you. Contact us. Every plan comes with the same safeguards.
Teams running Xevyo inside their own perimeter get seat-based billing, custom token budgets, SSO, and an SLA. That's what Enterprise is for.
Xevyo