Plans & Usage

Every organization in Rock8Cloud is on a plan that determines how many resources it can use and is billed via Stripe. You can view your current plan, manage your subscription, and check usage on the Settings > Billing page.

Available Plans

• Resource Pool - the CPU and memory your whole organization shares. Every running service draws its allocation from it
• Default Service Size - what a new service is allocated until you resize it
• Replicas per Service - how many copies of a service can run behind the same URL
• Build Minutes - total time spent building container images from your Dockerfiles
• AI Budget - what you can spend on AI features per period, see AI budget and spend
• Custom Domains - domains you can attach to each service
• Object Storage - total capacity across all S3 buckets in the organization
• Image Registry - storage for the images built from your code, and how many past builds of each service stay available for rollback
• Password-protected URLs - HTTP basic auth in front of a service URL, for previews and internal tools

Neptune (60 vCPU + 60 GiB pool, 2 replicas per service, €200 AI budget) and Enterprise are sales-assisted plans, see Enterprise.

All plans include a free trial. No payment information is required upfront.

New organizations automatically start on the Mercury plan with a free trial.

The Resource Pool

Your plan includes a pool of CPU and memory. Every awake service consumes its allocation from that pool. The rules:

• Every service has an allocation. New services start at 0.25 vCPU / 256 MiB on every plan, databases and one-click applications included. A blueprint can pin a larger starting size for one of its services. The size is stored on the service from that moment on and you can change it any time.
• Allocations are editable per service. On the service's Resources tab, resize in steps of 0.25 vCPU and 256 MiB, up or down (never below one step, at most 16 vCPU / 16 GiB per replica - a single pod still has to fit one machine, so bigger pools are spent through replicas and more services). Changes apply to the running pods immediately.
• Changing plans changes the pool, not your services. After an upgrade every service keeps the size it had. If a service needs more, raise it on its Resources tab. Before a downgrade, the total of your running services must fit the smaller pool.
• Replicas multiply consumption. A 0.5 vCPU service with 2 replicas draws 1 vCPU from the pool.
• Sleeping is free. A sleeping service consumes nothing. Sleep rarely-used services to make room for new ones.
• S3 buckets don't count. Object storage is metered by its own quota, not the pool.

When a deploy doesn't fit in the remaining pool, it isn't rejected. The service is parked with a clear reason and deploys automatically as soon as resources free up (you sleep, shrink, or delete something, or buy more capacity). Preview environments always yield to your stable services.

Buying More CPU and Memory

You can grow the pool on any plan with fixed-price monthly add-ons. One unit is 0.25 vCPU or 256 MiB, and units are cheaper on higher plans:

• You choose fixed quantities on the Billing page and can change them anytime. There is no metering and no usage-based billing - the price is the quantity you picked, nothing else.
• Add-on units carry no annual discount. The unit price is the same on monthly and annual billing, only the plan itself is discounted.
• Reducing add-ons is only possible when your current consumption still fits the smaller pool.

Free Trial

When you create a new organization, you get a free trial on the Mercury plan. During the trial:

• All Mercury plan features are available
• No payment method is required for the first 7 days
• Add a payment method and the trial extends to 30 days from signup, billing starts only when the trial ends
• You can upgrade to a higher plan at any time

After the trial ends, you'll need to add a payment method to continue using Rock8Cloud. If you don't subscribe, your organization enters read-only mode.

Subscription Management

Go to Settings > Billing to manage your subscription:

• Subscribe - Choose a plan and complete checkout via Stripe
• Switch Plans - Upgrade or downgrade between Pluto, Mercury and Earth
• Resource Add-ons - Buy or reduce extra CPU and memory units
• Manage Billing - Access the Stripe billing portal to update payment methods, view invoices, or cancel
• Monthly/Annual Toggle - Save 10% with annual billing

Upgrading

You can upgrade at any time. The new pool takes effect immediately. Stripe prorates the charge for the remaining billing period. Your usage counters (build minutes, AI spend) are preserved.

Downgrading

A downgrade is accepted only when your current consumption fits the smaller plan's pool (including any add-ons you keep). If your services use more CPU or memory than the target pool provides, shrink, sleep, or delete services first. The billing page tells you exactly which dimension is over.

Read-Only Mode

When your subscription is canceled or expired, your organization enters read-only mode:

• You can view dashboards, logs, and configurations
• You cannot deploy, build, create services, or run AI agents
• A warning banner appears on the billing page and in the navigation

To exit read-only mode, subscribe to any plan from the billing page.

AI Budget and Spend

AI features are metered in euros, not in runs. Every code review, code analysis, agent session, and custom agent run charges what the underlying model actually cost against your plan's included budget for the period.

Open Settings > AI to see:

• Enable AI Features - the master switch for the organization. Off means no AI feature runs at all
• Budget - spend against the included budget for this period, with the reset date
• Spend breakdown - over time, by model, by feature, and by project, for the last 24 hours, 7 days, or 30 days

When the included budget runs out, use Top up AI budget on the Billing page. €1 buys €1 of AI usage, from €5 up to €10,000 per purchase. Top-up balance never expires and is only spent after the included budget for the period is gone.

Usage Period

Usage counters (build minutes, AI spend) reset automatically every 30 days. The reset date is shown on the Billing page. Pool consumption, custom domains, and storage are always live - they reflect what currently exists, not a cumulative counter.

What Happens When You Hit a Limit

When a limit is reached:

• Resource pool - new deploys and preview environments are parked instead of started, with a banner in the deployment panel. Running services are never touched. Parked services deploy automatically once resources free up.
• Build minutes - new builds are rejected. Redeployments of already-built images still work.
• AI budget - AI code review, code analysis, and agents stop starting new runs until the period resets or you buy a top-up. Runs already in flight finish.
• Object storage - you cannot create a new bucket or raise a bucket's cap beyond what is left of the quota.

A badge on the dashboard shows which limits are exhausted, with a link to the Billing page.

Enterprise

For organizations that outgrow Earth, contact us about Neptune (60 vCPU + 60 GiB pool, 2 replicas per service, 10,000 build minutes, €200 AI budget per period) or our Enterprise tier with unlimited resources and custom support. Both are managed separately from the self-service billing system.