|
|
Posted on 3/25/2025 2:39:14 PM
|
|
|
|
Demand: With the explosion of AI large models, vector databases have also appeared in everyone's field of vision. Previously, we briefly introduced several vector databases, and this article tried to use Qdrant vector database storage and retrieval.
Qdrant vector database
Qdrant is an open-source vector database designed for next-generation AI applications. It is cloud-native and provides RESTful and gRPC APIs to manage embeddings. Qdrant has powerful features, supporting image, voice, and video search, as well as integration with AI engines.
Source:The hyperlink login is visible.
Download:The hyperlink login is visible.
Documentation:The hyperlink login is visible.
WebUI source code:The hyperlink login is visible.
API Documentation:The hyperlink login is visible.
Windows installs the Qdrant vector database
The official documentation tutorial is to deploy the Qdrant vector database directly with Docker, since I don't have the Docker environment installed natively, and then it feels faster to run it directly.
Download for Windows:The hyperlink login is visible.
WebUI Download:The hyperlink login is visible.
Regarding the 404 issue of accessing the dashboard /dashboard
Issue:Web UI is only included by default when using Docker images, if you want to use it with binaries, you will have to install the web UI file yourself.
Solution: Download the WebUI release package, create a new static folder under the qdrant directory, and then copy the files in dist into it.
The hyperlink login is visible.
The hyperlink login is visible.
Double-click to run“qdrant.exe”As shown below:
REST API: localhost:6333
Web UI: localhost:6333/dashboard
GRPC API: localhost:6334
Configuration Reference:The hyperlink login is visible.
Qdrant test
useBAAI/bge-m3Embed the model (which is 1024 dimensions), get the vector, and create a new collection in Qdrant for testing, as shown in the following figure:
Insert two pieces of data, as shown in the figure below:
The identity of the qdrantSupport for using 64-bit unsigned integers and UUIDs as identifiers for points。
Through vector search, the content is "Xiaohong likes programming, he likes to use .NET technology", as shown in the figure below:
The scores are: "score": 0.65278614, "score": 0.29873508, the closer to 1, the more matched.
C# calls the Qdrant vector database
You can install the Qdrant.Client library and use C# to read and write to Qdrant, with the following reference:
Code:
Reference:
The hyperlink login is visible.
The hyperlink login is visible. |
Previous:【AI】(14) A brief introduction to open source vector databasesNext:toPlainString, toEngineeringString, toString for BigDecimal in Java
|
Landlord