What you need first
- A running Quivr. On your machine,
make devstarts one; see the Quickstart. For a server, start with Run Quivr on one machine or Deploy and configure Quivr. - Its configuration file, the JSON file named by
QUIVR_CONFIG, and the right to restart theapiandworkerprocesses. Plugins, API keys and most limits are set there. - An operator API key. Runtime plugin changes, backfills and vector space promotions need
plugins:admin; following their progress needsoperations:read. Each page lists the exact permissions it uses. The API keys reference lists every permission. curlandjqfor the requests on these pages.
api and worker processes restart. Registering or activating a plugin, filling a vector space, reprocessing documents and retiring alert evaluations are API requests that apply while Quivr runs.
Install and upgrade
- Run Quivr on one machine: the quickest installation, from published images with one Compose file.
- Deploy and configure Quivr: what a deployment runs, the order to start it in, and how to check it is ready.
- Run Quivr behind TLS and Security of Quivr releases: encrypt traffic and verify what you install.
- Upgrade Quivr: move to a new release and keep a way back.
- Reset and reload an installation: rebuild an installation from your saved declarations, without setting up the database or sources by hand.
- Configuration: every field of the configuration file.
Plugins
Plugins are separate HTTP services that read files, cut and embed passages, rank results and judge alerts. How plugins work explains their roles.- First-party plugins: the plugins that ship with Quivr and their settings.
- Pin a plugin: make Quivr call a plugin at startup.
- Upgrade or switch a plugin: replace a plugin while Quivr runs, and roll back.
- Reprocess quarantined documents: run the documents a plugin refused or failed on again, once it is fixed.
Sources
- Receive and secure pushes: let a third party send records to Quivr, with its own token and limits.
Search
- Search profiles: what
defaultanddeepmean, and how to choose the plugin behind each. - Choose an embedding model: pick the model that computes vectors for meaning search.
- Re-rank deep searches with Jev: add an optional paid re-ranker to
deep. - Try and switch a vector model: compare a second embedding model, then switch search to it.
- Fill a new vector space: compute vectors for documents you already have.
Alerts
- Configure alert fields: name metadata fields that keyword alerts can filter on.
- Upgrade an alert rule: move existing alerts to a new version of their rule.
- Move meaning alerts to a new vector model: keep alerts that match by meaning working after search changes vector space.
- Retire alert evaluations: abandon evaluations whose old rule cannot come back.
Scale
- Scale Quivr: find capacity limits, budget large imports and protect live traffic.
- Run a large import: load an archive without starving live collection.
- Run the vector index on a dedicated host: give the search index its own machine.
Measure search (advanced)
These pages are for teams tuning search quality and capacity. You do not need them to run Quivr.- Record search measurements: store evaluation results and compare runs.
- Run local load tests: measure throughput and delays with free synthetic providers.
- Run a search campaign: explore search settings over several days, within a spending cap.
Next
- Deploy and configure Quivr if you have no deployment yet.
- Pin a plugin to add your first plugin.