vectrixdb 1.1.0

Where vectors come alive - A lightweight, visual-first vector database with embedded ML models

VectrixDB

Where vectors come alive.

A lightweight, visual-first vector database with embedded ML models - no API keys required.

Why VectrixDB?

Feature	VectrixDB	Qdrant	Chroma	Pinecone
Beautiful Dashboard	Yes	Basic	No	No
Embedded ML Models	Yes	No	No	No
4 Search Tiers	Yes	No	No	No
GraphRAG Built-in	Yes	No	No	No
Zero Config	Yes	No	Yes	Yes
No API Keys Needed	Yes	Yes	No	No
Open Source	Yes	Yes	Yes	No

Quick Start

pip install vectrixdb

from vectrixdb import Vectrix

# Create database with hybrid search (uses bundled English models)
db = Vectrix("my_docs", tier="hybrid", language="en")

# Add documents
db.add([
    "Python is great for data science",
    "JavaScript powers the web",
    "Rust is known for memory safety"
])

# Search
results = db.search("programming languages")
print(results.top.text)  # Best match

4-Tier System

Tier	Features	Use Case
dense	Vector similarity	Fast semantic search
hybrid	+ BM25 sparse	Better keyword matching
ultimate	+ ColBERT late interaction	Maximum accuracy
graph	+ Knowledge graph	Complex reasoning (GraphRAG)

# Dense tier (fastest)
db = Vectrix("docs", tier="dense")

# Hybrid tier (balanced)
db = Vectrix("docs", tier="hybrid")

# Ultimate tier (best quality)
db = Vectrix("docs", tier="ultimate")

# Graph tier (GraphRAG)
db = Vectrix("docs", tier="graph")

Search Modes

# Dense - vector similarity
results = db.search("AI", mode="dense")

# Sparse - BM25 keyword
results = db.search("machine learning", mode="sparse")

# Hybrid - combined
results = db.search("neural networks", mode="hybrid")

# Rerank - with cross-encoder
results = db.search("deep learning", mode="rerank")

With Metadata

db.add(
    texts=["iPhone 15", "Galaxy S24", "Pixel 8"],
    metadata=[
        {"brand": "Apple", "price": 999},
        {"brand": "Samsung", "price": 899},
        {"brand": "Google", "price": 699}
    ]
)

# Filter by metadata
results = db.search("smartphone", filter={"brand": "Apple"})

Embedded Models

VectrixDB bundles English models (~386MB) - no downloads needed:

Model	Purpose	Size
e5-small-v2	Dense embeddings	129MB
ms-marco-MiniLM	Reranking	129MB
answerai-colbert-small	Late interaction	129MB
BM25 vocab	Sparse search	17KB

Multilingual Models (auto-download)

For 100+ languages, models download from GitHub on first use:

# Multilingual (downloads ~450MB on first use)
db = Vectrix("docs", tier="hybrid")  # or language="multi"

# English only (bundled, no download)
db = Vectrix("docs", tier="hybrid", language="en")

Model	Purpose	Languages
multilingual-e5-small	Dense	100+
mmarco-mMiniLMv2	Reranking	15+
BGE-M3	Late interaction	100+
mREBEL	GraphRAG extraction	18

REST API & Dashboard

# Start server
VECTRIXDB_API_KEY=your_key vectrixdb serve --port 7337

# Open dashboard
# http://localhost:7337/dashboard

# Create collection
curl -X POST http://localhost:7337/api/v1/collections \
  -H "api-key: your_key" \
  -d '{"name": "docs", "dimension": 384}'

# Add with auto-embedding
curl -X POST http://localhost:7337/api/v1/collections/docs/text-upsert \
  -H "api-key: your_key" \
  -d '{"points": [{"id": "1", "text": "Hello world"}]}'

# Search
curl -X POST http://localhost:7337/api/v1/collections/docs/text-search \
  -H "api-key: your_key" \
  -d '{"query_text": "greeting", "limit": 10}'

Project Structure

VectrixDB/
├── vectrixdb/
│   ├── core/           # Vector index, storage, search
│   │   ├── graphrag/   # Knowledge graph
│   │   └── search/     # Search algorithms
│   ├── api/            # FastAPI server
│   ├── models/         # Embedded ONNX models
│   │   └── data/       # Bundled English models
│   ├── dashboard/      # Web UI
│   └── cli.py          # Command line
├── tests/              # Jupyter notebooks
└── requirements.txt

Installation from Source

git clone https://github.com/knowusuboaky/VectrixDB.git
cd VectrixDB
pip install -e .

Requirements

• Python 3.9+
• No external API keys
• Models bundled or auto-downloaded

License

Apache 2.0

Author

Kwadwo Daddy Nyame Owusu - Boakye

GitHub: @knowusuboaky

---

Where vectors come alive.