OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Vectordb — doganarif

Vectordb — doganarif is a self-hosted Python in-memory vector database with a FastAPI REST interface and a Python SDK. It stores embeddings in libraries, documents and chunks, and supports similarity search, metadata filtering, and linear, KD-tree or LSH indexing.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Vectordb — doganarif — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an application already has embedding vectors and needs an in-memory retrieval service with a Python client or REST interface.

Who it is for

  • Developers adding vector retrieval to Python applications or RAG systems.
  • Teams that want to expose vector storage and search through a self-hosted REST service.

Documented uses

  • Store text chunks with embeddings and metadata.
  • Build an index and search for similar vectors using a query embedding.
  • Filter search results by metadata.
  • Choose among documented indexing algorithms and supported similarity metrics.
Before you choose

Requirements & limits

The project is published as a public GitHub repository. Its README documents Docker and local Python setup.

  • Index and metric support are coupled: linear supports cosine and Euclidean; KD-tree supports Euclidean; LSH supports cosine.
  • The default repository is in-memory and per process; the README warns that multiple workers keep separate data copies and snapshots.
  • The README lists snapshot and restore functionality and describes file-based, database-backed or external shared storage as alternatives for persistence or multi-worker use.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a local vector search and run the documented example

Start the service locally, then use the Python SDK example to add vectors, build an index and search.

Before you start

  • Python and the repository's requirements file for local setup
  • Repository source at https://github.com/doganarif/vectordb

Steps

  1. Create and activate a Python virtual environment.

  2. Install dependencies with the command below.

    Documented command
    pip install -r requirements.txt

  3. Start the service with the command below or the documented Uvicorn command.

    Documented command
    make run

  4. Use VectorDBClient to create a library, document and chunks with embeddings, then build an index and search with a query vector.

Expected outcome

A local FastAPI service and a Python SDK workflow for storing vectors and retrieving similar results.

Guide limitations

  • The documented data repository is in-memory and per process; multiple workers maintain separate data copies.
  • The example supplies embedding vectors. Cohere embedding support is listed separately and requires a configured API key when used.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
doganarif/vectordb · Official homepage title · retained in its source language
How we review sourcesSuggest a correction