Configuration
Vectra uses environment variables for its configuration. Most of these can be set in a .env file at the root of the project.
Core Settings
| Variable | Description | Default |
|---|---|---|
ENV |
Environment (development, production) | development |
SECRET_KEY |
Secret key for JWT authentication | (Required) |
DATABASE_URL |
SQLAlchemy URL for the PostgreSQL database | (Required) |
QDRANT_HOST |
Hostname for the Qdrant vector database | localhost |
REDIS_HOST |
Hostname for the Redis cache/queue | localhost |
AI Providers
Vectra supports multiple LLM and Embedding providers:
- Gemini:
GEMINI_API_KEY - OpenAI:
OPENAI_API_KEY - Mistral:
MISTRAL_API_KEY - Ollama:
OLLAMA_BASE_URL(usuallyhttp://localhost:11434)
Data Persistence (Docker)
To allow Vectra to access your local folders when running in Docker, you must define the mapping:
# The physical path on your host machine
VECTRA_DATA_PATH=D:/MyDocuments
Any Connector path starting with this value will be correctly mapped to the internal Docker /data volume.
Initial Setup
On first startup, Vectra can automatically create a superuser:
FIRST_SUPERUSER=admin@vectra.ai
FIRST_SUPERUSER_PASSWORD=secure_password_here
[!TIP] For a complete list of available settings and their meanings, please refer to the app/core/settings.py file.