# LLM access

A Project's LLM allocation determines its available models, shared budget and
rate limits. Open the Project dashboard to inspect your allocation and available
chat and API access.

## Open chat

Use the Project's chat entry and complete sign-in if requested. On a first visit,
the chat account may need to be created before the platform can apply its
membership. A preparing allocation may therefore need time before its models
appear.

Choose one of the models granted to the Project. The available names depend on
the allocation and the installation's model catalogue.

## Read your budget

The dashboard shows the approved monthly budget, measured spend and remaining
amount in EUR, together with the measurement time and reset information.
Development installations may use virtual accounting prices.

Usage can arrive after the request finishes. An unavailable measurement is shown
as unknown; a measured zero means a successful measurement found no spend.
Requests per minute and tokens per minute are separate rate limits.

## Request a personal API key

Use the personal-key control for an eligible allocation. The platform emails the
key to your verified address; the dashboard shows delivery status and never
reveals the key itself. A pending request may need time for provisioning and mail
delivery.

Use the gateway address supplied by the installation and a model permitted by
your allocation. Store the key in your client's secret settings or environment,
and keep it out of code, screenshots and support tickets.

Revoke the key from the platform when it is no longer needed or may have been
exposed. Revocation also cancels pending delivery. Access remains subject to the
allocation's limits and current membership policy.

If chat or key delivery remains unavailable, include the Project and the visible
status in a [support request](support.md).