llamate 0.1.0

A memory-augmented framework for LLMs

🦙 Llamate

A memory-augmented agent framework for LLMs.

Quick Start

1. Install Llamate

pip install llamate

2. PostgreSQL Setup

To use Llamate with PostgreSQL:

# Run PostgreSQL with pgvector extension using Docker
docker run --name llamate-postgres -e POSTGRES_USER=llamate -e POSTGRES_PASSWORD=llamate -e POSTGRES_DB=llamate -p 5432:5432 -d ankane/pgvector

3. Verifying PostgreSQL Installation

Check if the container is running:

docker ps | grep llamate-postgres

Check PostgreSQL logs:

docker logs llamate-postgres

4. Accessing PostgreSQL Container

Connect to PostgreSQL:

docker exec -it llamate-postgres psql -U llamate -d llamate

Useful PostgreSQL Commands:

-- List all tables
\dt

-- Check if pgvector extension is installed
\dx

-- List all schemas
\dn

-- Exit PostgreSQL shell
\q

5. Initialize Llamate config

llamate --init

When prompted, choose postgres as your vector backend and enter this connection string:

postgresql://llamate:llamate@localhost:5432/llamate

6. Basic Usage

from llamate import MemoryAgent, get_vectorstore_from_env

# Create an agent with your config settings
user_id = "test_user"
vectorstore = get_vectorstore_from_env(user_id=user_id)
agent = MemoryAgent(user_id=user_id, vectorstore=vectorstore)

# Add a memory
response = agent.chat("The capital of France is Paris")

# Retrieve memory later
response = agent.chat("What's the capital of France?")
print(response)  # Will include information about Paris

### 7. Interactive CLI

```bash
llamate

Complete PostgreSQL End-to-End Flow

Follow these steps to set up, use, and view data in Llamate with PostgreSQL:

1. Start PostgreSQL Container

docker run --name llamate-postgres -e POSTGRES_USER=llamate -e POSTGRES_PASSWORD=llamate -e POSTGRES_DB=llamate -p 5432:5432 -d ankane/pgvector

2. Initialize Llamate

llamate --init
# Select 'postgres' as your vector store backend
# Enter connection string: postgresql://llamate:llamate@localhost:5432/llamate

3. Run a Test Script to Store Data

Create a file test_llamate.py:

from llamate import MemoryAgent, get_vectorstore_from_env
import os

# Set user ID
user_id = "test_user"

# Initialize components
vectorstore = get_vectorstore_from_env(user_id=user_id)
agent = MemoryAgent(user_id=user_id, vectorstore=vectorstore)

# Add memories
agent.chat("The capital of France is Paris.")
agent.chat("The Eiffel Tower is 324 meters tall.")
agent.chat("Python is a programming language created by Guido van Rossum.")

# Test retrieval
response = agent.chat("Tell me about Paris.")
print("Response:", response)

4. View Data in PostgreSQL

Connect to the database:

docker exec -it llamate-postgres psql -U llamate -d llamate

List tables to find your memory table (it will use your user_id):

\dt

View table structure:

\d memory_test_user

Display memory records (omitting the large vector field):

SELECT id, text FROM memory_test_user;

Count records:

SELECT COUNT(*) FROM memory_test_user;

Query specific memories (using text search):

SELECT id, text FROM memory_test_user WHERE text LIKE '%Paris%';

Delete test memories (if needed):

DELETE FROM memory_test_user WHERE text LIKE '%test%';

Exit the PostgreSQL shell:

\q

Features

• Persistent memory for AI using vector embeddings
• Multiple vector store backends (FAISS and PostgreSQL)
• Easy integration into existing applications
• Simple CLI for testing and demonstration