Skip to content

Configuration

Configure Pygrad's LLM provider, embeddings, and storage.

Overview

Pygrad uses Cognee under the hood, which supports various LLM providers and databases. Configuration is done through environment variables.

Ollama (Local)

Run with local LLMs for privacy and offline use.

OpenAI

Use OpenAI's GPT models and embeddings.

Database

Configure PostgreSQL, Neo4j, and other backends.

Environment Variables

All configuration is done through environment variables. You can set them directly or use a .env file.

Core Variables

Variable Description Default
LLM_PROVIDER LLM provider name openai
LLM_MODEL Model name gpt-4o
LLM_API_KEY API key -
LLM_ENDPOINT Custom endpoint URL -
EMBEDDING_PROVIDER Embedding provider openai
EMBEDDING_MODEL Embedding model name text-embedding-3-small
EMBEDDING_ENDPOINT Custom embedding endpoint -
EMBEDDING_DIMENSIONS Vector dimensions 1536

Storage Variables

Variable Description Default
VECTOR_DB_PROVIDER Vector database lancedb
DB_PROVIDER Relational database sqlite
GRAPH_DATABASE_PROVIDER Graph database networkx

Configuration File

Create a .env file in your project directory:

# LLM Configuration
LLM_PROVIDER="ollama"
LLM_MODEL="qwen3-coder:30b"
LLM_API_KEY="ollama"
LLM_ENDPOINT="http://localhost:11434/v1"

# Embedding Configuration
EMBEDDING_PROVIDER="ollama"
EMBEDDING_MODEL="embeddinggemma:latest"
EMBEDDING_ENDPOINT="http://localhost:11434/api/embed"
EMBEDDING_DIMENSIONS="768"

# Optional
TELEMETRY_DISABLED=true

Loading Configuration

Pygrad automatically loads environment variables. For Python scripts:

from dotenv import load_dotenv
load_dotenv()  # Load from .env file

import pygrad as pg
# Now pg.add(), pg.search(), etc. will use your configuration

Next Steps