Join us in San Francisco for TailscaleUp!Grab your ticket ->

Aperture dashboard reference

Last validated:
Aperture by Tailscale is currently in beta.

Access the Aperture dashboard at /ui/ on your Aperture hostname (for example, http://<aperture-hostname>/ui/).

The dashboard is organized into groups in the left sidebar. This page documents the pages every user can access, under the My Aperture and Usage & Adoption groups. For the admin-only pages (Security, Administration, Reference, and the admin pages within Usage & Adoption), refer to the Aperture admin dashboard reference, which covers content visible only to users with the admin role.

This topic is part of the Aperture reference documentation. For how-to guides on observing and exporting usage data, refer to Observe and export AI usage.

My Aperture

The My Aperture group holds the pages every user can access for their own activity and configuration.

My Dashboard

If you incorrectly authenticate user devices with tag identities, all LLM sessions from those devices appear to come from the tag identity instead of the individual user. This happens because Tailscale tags do not provide per-user identity. Refer to all sessions appear to come from the same user for more information.

The My Dashboard page shows token usage for the current user. It includes the following components:

  • Total tokens by model (bar chart)
  • Recent activity timeline
  • Quick stats: total requests, total tokens, active sessions

When header filtering is configured, you can also filter this page by indexed request and response headers.

Admins can access other users' dashboards by going to /ui/dashboard/:loginName where loginName is the Tailscale login name of the user (for example, alice). For instructions on checking and managing budgets, refer to check and refill budgets.

Agent Config

The Agent Config page provides setup instructions for connecting an LLM client or agent through Aperture. The instructions adjust based on whether any providers are configured yet, so you can point a client at your Aperture endpoint and start routing requests.

Models

The Models page lists configured models grouped by provider:

  • Model ID and provider
  • Inline pricing per million tokens (input and output), with an expandable tooltip for cache, reasoning, image, fast-mode pricing, and quotas
  • A test button on each row that runs an instant API test against the model. The button is disabled for a model whose provider is configured in passthrough mode with no fallback apikey, because the test runs from the Aperture device and has no client credential to forward.

Users can reference this page to choose which model to use through Aperture when configuring LLM clients. For a list of supported providers and API formats, refer to supported models and providers.

Connectors

The Connectors page lists tools and integrations for your AI agents. For non-admin users, Aperture omits configured connectors they cannot discover. Admins can access all configured connectors, and system connectors remain visible with their access status.

Select a connector to review its status. For an accessible MCP connector, the connector details also list tools, resources, and resource templates when capability data is available. A connector that uses per-user OAuth can ask you to authorize or reconnect your account. Built-in connectors can provide guided setup actions when the applicable experimental feature is enabled.

Connector grants determine which tools, resources, templates, and HTTP proxy capabilities you can use. For setup, status, and authorization details, refer to Use a connector.

Admins can use Manage to open the admin connector configuration. For the admin page, refer to Connectors in the admin dashboard reference.

Chat

The Chat link is available when chat is enabled for your account on the Aperture gateway.

The Chat link opens the Aperture chat application in a new tab. It is an external application outside the dashboard, so it opens at /chat/ on your Aperture hostname rather than within the /ui/ dashboard.

Usage & Adoption

The Usage & Adoption group holds organization-wide usage data. The Usage Overview page is visible to every user. The remaining pages in this group are admin-only and are documented in the Aperture admin dashboard reference.

Usage Overview

The Usage Overview page shows organization-wide adoption metrics:

  • A daily aggregate chart of cost or tokens over time.
  • Stat cards for active users, total tokens, and estimated cost, with a token-type breakdown.
  • A sortable, paginated table of top users showing rank, login name, token counts (input, output, cached, reasoning, total), and average cost per million tokens.

When header filtering is configured, you can also filter this page by indexed request and response headers.

This page helps platform teams understand LLM adoption patterns and identify users who need support or training.

For non-admin users, other users' cost fields are redacted, so each user can access only their own cost. The Download CSV button, which exports usage data including token counts per model, user, and date, is available only to admins. To export this data for external analysis, refer to Export usage data to S3.