> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rafftechnologies.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Account limits

> Per-account quotas that govern how many VMs and how much CPU/RAM you can run at once

<sub>Updated May 8, 2026</sub>

Every account has soft caps on how many VMs and how much aggregate CPU and RAM it can hold simultaneously. These limits exist to keep the platform fair, prevent abuse (runaway scripts, mining), and give billing a safety net.

## What's limited

| Limit        | Default for new accounts | Counts                                         |
| ------------ | ------------------------ | ---------------------------------------------- |
| **VM count** | **30 VMs**               | Total VMs across `active` and `passive` states |
| **vCPU sum** | Tied to plan choices     | Total vCPU across all your VMs                 |
| **RAM sum**  | Tied to plan choices     | Total RAM (GB) across all your VMs             |

New accounts start with a **30-VM default**. Quotas increase automatically as you build a usage history, and can be raised on request — see *Request a limit increase* below.

## When you hit a limit

Creation or resize-up returns a clear error naming the limit that was hit (`too many VMs`, `RAM quota exceeded`). Your options:

1. **Delete VMs you no longer need** — including stopped (`passive`) ones, since they still count.
2. **[Resize](/products/build/virtual-machines/quickstart-guides/resize) down** — drop CPU/RAM on existing VMs to make room for the new one.
3. **Request a limit increase** — see below.

## Stopped VMs still count

A `passive` VM holds its CPU, RAM, disk, and IP reservations on the hypervisor and counts the same as a running VM toward your quota. Stopping a VM does not free quota — only **deleting** it does (unused subscription time is credited back to your balance).

## Request a limit increase

Quotas can be raised on request. Standard turnaround is **24 hours** for limit-increase emails.

* **Standard increases** — email [support@rafftechnologies.com](mailto:support@rafftechnologies.com) with your account ID, the limit you're hitting, and the workload you're planning. Response within **24 hours** in business hours; reasonable bumps are typically approved on the same response.
* **Urgent increases** — if you're actively launching, in the middle of an incident, or have a time-sensitive workload, **mark the request as urgent in the subject** and include the account ID and the deadline; support can lift quotas in real time when context warrants it.
* **Planned bursts** — load tests, conferences, batch jobs, traffic events: give support a heads-up before the window so they can pre-stage the increase.

What to put in the request:

| Field             | Example                                         |
| ----------------- | ----------------------------------------------- |
| Account ID        | from the dashboard's account dropdown           |
| Current limit     | "30 VMs / 250 GB RAM"                           |
| Requested limit   | "60 VMs / 500 GB RAM"                           |
| Why               | "Spinning up a CI fleet for a 2-week migration" |
| Deadline (if any) | "Need by Monday — production cutover"           |

## Related

<CardGroup cols={2}>
  <Card title="Plans & sizing" icon="lightbulb" href="/products/build/virtual-machines/concepts/plans-and-sizing">
    What CPU and RAM each plan reserves.
  </Card>

  <Card title="Lifecycle states" icon="lightbulb" href="/products/build/virtual-machines/concepts/lifecycle-states">
    Why `passive` keeps consuming quota.
  </Card>
</CardGroup>