Search

Explore Qdrant’s search capabilities, from basic similarity search to advanced hybrid and multimodal queries. These pages cover how to find relevant results, apply filters, combine query types, and tune search quality and latency.

Search

Search describes similarity search — finding points whose vectors are closest to a query vector in the configured vector space.

Filtering

Filtering lets you narrow results using payload conditions, combining database-style clauses with vector search for precise retrieval.

Hybrid Queries

Hybrid Queries combine multiple queries or execute them in stages, enabling fusion of dense and sparse results and complex re-ranking pipelines.

Explore

Explore covers discovery APIs such as recommendations and random sampling for browsing and navigating collections.

Text Search explains Qdrant’s built-in full-text search capabilities and how to combine them with vector search.

Search Relevance

Search Relevance describes techniques to improve result ranking beyond raw vector similarity, including score boosting and re-ranking strategies.

Low-Latency Search covers configuration and design patterns for achieving the fastest possible query response times.

Was this page useful?

Thank you for your feedback! 🙏

We are sorry to hear that. 😔 You can edit this page on GitHub, or create a GitHub issue.