Qdrant (read: quadrant ) is a vector similarity search engine. It provides a production-ready service with a convenient API to store, search, and manage points - vectors with an additional payload.

This CLI starts a Qdrant peer/server.

Usage: qdrant [OPTIONS]

Options:
      --bootstrap <URI>
          Uri of the peer to bootstrap from in case of multi-peer deployment. If not specified - this peer will be considered as a first in a new deployment

          [env: QDRANT_BOOTSTRAP=]

      --uri <URI>
          Uri of this peer. Other peers should be able to reach it by this uri.

          This value has to be supplied if this is the first peer in a new deployment.

          In case this is not the first peer and it bootstraps the value is optional. If not supplied then qdrant will take internal grpc port from config and derive the IP address of this peer on bootstrap peer (receiving side)

          [env: QDRANT_URI=]

  -f, --force-snapshot
          Force snapshot re-creation If provided - existing collections will be replaced with snapshots. Default is to not recreate from snapshots

      --snapshot <PATH:NAME>
          List of paths to snapshot files. Format: <snapshot_file_path>:<target_collection_name>

          WARN: Do not use this option if you are recovering collection in existing distributed cluster. Use `/collections/<collection-name>/snapshots/recover` API instead.

      --storage-snapshot <PATH>
          Path to snapshot of multiple collections. Format: <snapshot_file_path>

          WARN: Do not use this option if you are recovering collection in existing distributed cluster. Use `/collections/<collection-name>/snapshots/recover` API instead.

      --config-path <PATH>
          Path to an alternative configuration file. Format: <config_file_path>

          Default path: config/config.yaml

      --disable-telemetry
          Disable telemetry sending to developers If provided - telemetry collection will be disabled. Read more: <https://qdrant.tech/documentation/guides/telemetry>

      --stacktrace
          Run stacktrace collector. Used for debugging

      --reinit
          Reinit consensus state. When enabled, the service will assume the consensus should be reinitialized. The exact behavior depends on if this current node has bootstrap URI or not. If it has - it'll remove current consensus state and consensus WAL (while keeping peer ID) and will try to receive state from the bootstrap peer. If it doesn't have - it'll remove other peers from voters promote the current peer to the leader and the single member of the cluster. It'll also compact consensus WAL to force snapshot

  -h, --help
          Print help (see a summary with '-h')

  -V, --version
          Print version