Policies · 7 of 8

Service Level Agreement

Updated 2026-10-14

What we aim for

Lirran is meant to be open whenever the work needs it — a launch the next morning, a campaign going live, a client review in an hour. We run the platform around the clock and keep it available with commercially reasonable effort.

What depends on others

Every render is made by a third-party model provider. When a provider slows down, queues requests or pauses a model, that engine may slow down or pause here as well. Because their uptime is not ours to promise, we do not publish a fixed uptime percentage — we would rather say so plainly than print a number we cannot keep.

The remedy is built in

A render that fails, times out or is rejected by its provider returns its credits to your balance automatically. You never pay for a render that did not arrive. If one engine is having a bad day, the others usually are not — your balance works on all of them.

Render times

Images are usually quicker than video, but render time depends on the engine, the length you choose and the provider's queue. When a render runs long, the platform keeps checking on it and files it in your library the moment it is ready.

Maintenance

Planned maintenance is kept short and scheduled for quiet hours wherever possible. Renders already in progress are not lost; if one cannot finish, it is refunded.

Support

[email protected] answers within one business day, every business day. If the service is unavailable for an extended period, unused purchased credits remain refundable under the Refunds Policy, and the credits you hold never expire in the meantime.

Something unclear? A person reads every message.