Dashboard

User Management

Dashboard accounts. Admins manage users here and see the Health and Qdrant views.

Activity Summary

Key counts for the selected time window, across all users. The origin filter applies to the panels below, not to these cards.

Usage, ratings and reports

Ratings and reports count on the day their answer was sent, not the day the feedback arrived.

OpenRouter Credits

Balance and spend of this API key, not the whole account.

Status

Upstream providers, service probes and the latest observed events.

LLM Requests

Chat-completion calls since request tracing was enabled.

Health Incidents

Upstream LLM and embedding outages, last 30 days; repeats are grouped.

Model Deprecation

Provider-reported expiration date of every configured model role.

Storage

Postgres size and Qdrant on disk (incl. segment/WAL preallocation).

Activity by User

Top users ranked by total activity, with rating quality and error count.

Recent Questions

Each row pairs a question with its answer, rating, report, planner reasoning, and consulted documents.

Improvements

Open feedback and improvement ideas awaiting a fix. Add your own items with the plus button; mark an item as implemented once it is addressed. Resolved and dismissed items leave this list but stay readable under the status filter.

Qdrant Projection

2D UMAP projection of the collection's vectors, colored by content type. Hover a point for metadata; search to highlight matches.

Benchmark Cases

Run Benchmark

Each case runs against every listed model (up to 3), repeated per the configured repetitions, scored 0–10 (answer quality minus routing and document penalties). A model names an orchestrator and a retriever, which may differ; each role and the rater can also pick a reasoning effort (Default keeps the deployment's configured effort). One rater scores the whole run and defaults to the configured response rater. Repetitions and penalties live in config.yml. Runs execute in the backend; progress appears live below as a per-case results table.

Cases: All cases runs the working table above. Picking a collection instead runs that collection's saved set — not the rows shown above, which may have been edited since the collection was last saved.

Score Distribution

Per-model final scores (0–10) with one dot per scored unit colored by case category. Hover a dot for details, click to open the full result, toggle categories from the legend.

Run History

Past benchmark runs, newest first. Click a run to open its full per-case results. Tick another finished run over the same cases to show it alongside the open run, read-only. Delete removes the run and its results.

Digest

Ask about chatbot usage in plain language — answered live from the database, read-only.

Support Requests

Individual support requests users raised. A conversation may have several requests; each row is one request. Assign responsible users, open a request to read the full transcript, and mark it resolved. Add requests received by mail, phone or in person with the plus button.