- Stars
- 35.5k
- Forks
- 2.8k
- Licence
- Apache-2.0
- Language
- Rust
- Version
- 1.20.0
- Updated
At a glance
- Filtered vector search combining similarity and metadata
- Quantisation to reduce memory use
- REST and gRPC APIs with clients for major languages
Get started
Run with Docker
docker run -p 6333:6333 qdrant/qdrantWhat it is
Qdrant stores vectors, such as text or image embeddings, alongside JSON payloads, and finds the nearest matches to a query vector.
Why it stands out
Its strength is filtering: similarity search can be combined with conditions on the payload without losing speed.
Who it is for
Teams building semantic search, recommendations or retrieval for LLM applications.
