Qdrant Vector Search Engine
Build Semantic Search Tools on Qdrant Cloud
Send raw text or images to Qdrant Cloud, let Qdrant create the embeddings, and search by meaning in one API call. Start free with no credit card.
Search by meaning
Match Queries by Meaning When the Words Differ
Semantic search compares embeddings, numeric representations of meaning, so a query finds documents that say the same thing in different words.
Find similar images in a large catalog.
Embed images with a multimodal model such as CLIP, then search them with a text query or another image.
Keep metadata on every result.
Store a JSON payload next to each vector, so the record you filter on is the record you return.
Cloud Inference
Let Qdrant Cloud Create Embeddings for Your Semantic Search Engine
On Qdrant Cloud, Cloud Inference generates embeddings inside your cluster, so one system creates your vectors, stores them, and searches them.
Upsert raw text and get vectors back in one call.
Pass a document and a model name, and Qdrant embeds it on write and on query.
Choose dense, sparse, or multimodal models.
Use models hosted in Qdrant Cloud, or route requests to OpenAI, Cohere, or Jina.
Start on free models.
Several embedding models are free to use, including on free clusters.
Filtered search
Filter by Price or Date and Still Rank by Meaning
Most apps need to filter and rank at once. Qdrant applies payload filters, such as in stock or this year, while it searches the vectors.
Filter inside the graph.
Index a payload field, and Qdrant adds links between points that share its values, so filtered queries keep a connected path to the nearest neighbors.
Index filter fields early.
Create payload indexes right after you create the collection, before you ingest data.
Semantic code search
Search a Large Codebase in Natural Language
Index your codebase as vectors and search it by what the code does, such as "the code that retries a failed upload."
Query code in natural language.
Describe the logic you need, and get back the snippets that implement it.
Search by similar logic.
Start from a snippet you have and get back code that works the same way.
Results
What Teams Measured After Moving Search to Qdrant
Tripadvisor
over a billion reviews and images indexed on Qdrant Cloud, and users of its AI trip planning features show 2 to 3x more revenue.
Read the Tripadvisor case studyHimalayas
semantic and keyword search in one query: 25M+ requests a month at about 30 ms median inside Qdrant, and 19 points more recall after retrieval improvements.
Read the Himalayas case studyBazaarvoice
semantic search and recommendations over billions of vectors on Qdrant Hybrid Cloud: about 100x less storage than in PostgreSQL, with sub-100 ms queries at about 98% accuracy.
Read the Bazaarvoice case studyGetting started
Go From Sign-Up to First Semantic Query
Create a free cluster.
Sign up for Qdrant Cloud with no credit card.
Upload your data as text.
Let Cloud Inference embed it with a free model, and add payload fields you want to filter on.
Call the semantic search API.
Send a text query, and Qdrant embeds it and returns the closest matches by meaning.
Common Questions
How do I get started with Qdrant Cloud?
Do I need to host an embedding model?
Can I combine semantic search with keyword search?
