qdrant-haystack
A Python integration that lets the Haystack framework use Qdrant as its document store, as a drop-in replacement for other vector embedding stores.
qdrant-haystack integrates the Qdrant vector database with the Haystack framework by deepset, providing a QdrantDocumentStore class that can replace any other vector embeddings store in a Haystack application with minimal changes.
It installs from PyPI with pip or poetry, and its constructor accepts the same kinds of parameters as the official Qdrant Python client, including index name, embedding dimension and HNSW configuration. It supports Qdrant's local in-memory mode for tests and quick experiments, an on-disk local mode that persists vectors between runs without requiring a separate Qdrant server, and a networked Qdrant deployment for production use.
The README notes that this V2 implementation has moved to the deepset-ai/haystack-core-integrations repository, where the current documentation lives; this repository holds the earlier implementation. It is Apache-2.0 licensed.