Vector database
Open sourceQdrant for businesses
Qdrant is an open-source vector database that makes texts searchable according to their meaning instead of exact words. We use Qdrant as the memory of our AI document search and chatbots: It finds the text passages on the basis of which the language model responds in fractions of a second.
- Licence
- Apache 2.0
- Operation
- Self-hosted in Germany or at your premises
- Official website
- qdrant.tech
What we use Qdrant for
AI Document Search
Documents are broken down into sections and stored as vectors; Questions find the right places even with a different choice of words.
Chatbots with company knowledge
The chatbot only responds on the basis of the content found, released and can name the source.
Filtered search
Hits can be restricted by department, document type, date or authorization.
Combined search
Meaning search and keyword search can be combined so that article numbers and technical terms can also be found reliably.
Why Qdrant?
Apache 2.0 License
True open source license with no restrictions for commercial use.
Fast and economical
Qdrant is written in Rust and can also handle large document stocks on normal hardware.
Filter on metadata
Permissions and categories are taken into account directly in the search, not only afterwards.
Self-hostable
Runs as a container on your server; the vectors of your documents do not leave your infrastructure.
What we take care of with Qdrant
- Setting up and securing Qdrant (access protection, encryption)
- Preparation of your documents: disassembly, embedding, metadata
- Automatic update when documents are added or changed
- Tuning the search to find the right jobs
Implementation and consulting can often be subsidised through German BAFA or INQA funding.
What it is not meant for
A vector database is not a substitute for a document archive and not a database for accounting or customer data. It is a search index – the originals remain in filing or DMS.
Our solutions with Qdrant
Frequently asked questions about Qdrant
What is a vector database?
+−
A database that stores texts as series of numbers (vectors) that depict their meaning. For example, a search for "vacation entitlement" also finds a paragraph about "vacation entitlement," even though the word does not appear.
What do you need Qdrant for in an AI chatbot?
+−
So that the chatbot can access your knowledge: Qdrant finds the right text passages, and the language model formulates the answer from them. This process is called retrieval-augmented generation (RAG).
Is Qdrant free?
+−
Yes, the software is under the Apache 2.0 license. Costs are only incurred for the server and the setup. In addition, there is a paid cloud version from the manufacturer, which we usually do not use for sensitive data.
How up-to-date is the data in the search?
+−
As up-to-date as the connection intends. We set up new or changed documents to be automatically re-imported.
Go deeper
- AI consultancy
- Qdrant or pgvector? The right vector database for SMEs
- Hybrid search: Why an AI search must also find article numbers
- Permissions in AI search: So that everyone can only find what they are allowed to see
- Vector databases explained simply – using Qdrant as an example
- Building a RAG system in a company: process, costs, common mistakes
- RAG explained simply: How AI answers with your own documents
- RAG or Fine-Tuning? How AI really uses your company knowledge
Works together with
Product names and logos mentioned are trademarks of their respective owners. bettersorted uses this software and operates it for clients; there is no official partnership.
A 30-minute conversation, free of charge.
A 30-minute consultation, free of charge. We’ll take a look at your processes and give you our honest opinion on whether automation is worthwhile.