LLM Wrapper

A comprehensive Python wrapper for Large Language Models with database integration and usage tracking. Supports multiple database backends (PostgreSQL, MySQL, MongoDB) and provides detailed analytics for LLM usage.

Features

• 🚀 Easy Integration: Simple API for interacting with various LLM services
• 📊 Usage Tracking: Comprehensive logging and analytics for all LLM requests
• 💾 Multi-Database Support: PostgreSQL, MySQL, and MongoDB backends
• ⚡ High Performance: Optimized for concurrent requests and high throughput
• 🔒 Secure: Built-in security features and API key management
• 📈 Analytics: Detailed usage statistics and reporting
• 🐳 Production Ready: Robust error handling and logging

Installation

Basic Installation

pip install llm-wrapper

Development Installation

pip install llm-wrapper[dev]

With All Optional Dependencies

pip install llm-wrapper[dev,test,docs]

Quick Start

Basic Usage

from llm_wrapper import LLMWrapper

# Configure your database
db_config = {
    'type': 'postgresql',
    'dbname': 'llm_wrapper_db',
    'user': 'postgres',
    'password': 'your_password',
    'host': 'localhost',
    'port': '5432'
}

# Initialize the wrapper
wrapper = LLMWrapper(
    service_url="https://your-llm-service.com",
    api_key="your-api-key",
    db_config=db_config,
    deployment_name="gpt-4",
    api_version="2024-12-01-preview",
    default_model='gpt-4'
)

# Send a request
response = wrapper.send_request(
    input_text="What are the benefits of renewable energy?",
    customer_id=1,
    organization_id=1,
    temperature=0.7,
    max_tokens=2000
)

print(f"Response: {response['output_text']}")
print(f"Tokens used: {response['total_tokens']}")

# Get usage statistics
stats = wrapper.get_usage_stats()
print(f"Total requests: {stats['total_requests']}")
print(f"Total tokens: {stats['total_tokens']}")

# Clean up
wrapper.close()

Environment Variables

Create a .env file for easier configuration:

# Database Configuration
DATABASE_TYPE=postgresql
DB_NAME=llm_wrapper_db
DB_USER=postgres
DB_PASSWORD=your_password
DB_HOST=localhost
DB_PORT=5432

# LLM Service Configuration
LLM_SERVICE_URL=https://your-llm-service.com
LLM_API_KEY=your-api-key
LLM_DEPLOYMENT_NAME=gpt-4
LLM_API_VERSION=2024-12-01-preview

Database Configurations

PostgreSQL

db_config = {
    'type': 'postgresql',
    'dbname': 'llm_wrapper_db',
    'user': 'postgres',
    'password': 'your_password',
    'host': 'localhost',
    'port': '5432'
}

MySQL

db_config = {
    'type': 'mysql',
    'dbname': 'llm_wrapper_db',
    'user': 'root',
    'password': 'your_password',
    'host': 'localhost',
    'port': '3306'
}

MongoDB

db_config = {
    'type': 'mongodb',
    'dbname': 'llm_wrapper_db',
    'host': 'localhost',
    'port': 27017,
    'user': 'your_user',
    'password': 'your_password'
}

MongoDB with Connection String

db_config = {
    'type': 'mongodb',
    'dbname': 'llm_wrapper_db',
    'connection_string': 'mongodb://localhost:27017/'
}

Usage Analytics

# Get overall statistics
stats = wrapper.get_usage_stats()

# Get customer-specific statistics
customer_stats = wrapper.get_usage_stats(customer_id=1)

# Get organization-specific statistics
org_stats = wrapper.get_usage_stats(organization_id=1)

# Get statistics for a specific time period
from datetime import datetime, timedelta

start_date = datetime.now() - timedelta(days=7)
end_date = datetime.now()

period_stats = wrapper.get_usage_stats(
    start_date=start_date,
    end_date=end_date
)

Configuration Options

Parameter	Type	Default	Description
service_url	str	Required	LLM service endpoint URL
api_key	str	Required	API key for authentication
db_config	dict	Required	Database configuration
deployment_name	str	None	LLM deployment name
api_version	str	None	API version
default_model	str	'gpt-3.5-turbo'	Default model to use
timeout	int	30	Request timeout in seconds
max_retries	int	3	Maximum retry attempts
retry_delay	float	1.0	Delay between retries

API Reference

Core Methods

send_request(input_text, customer_id, organization_id, **kwargs)

Send a request to the LLM service.

Parameters:

• input_text (str): The prompt text
• customer_id (int): Customer identifier
• organization_id (int): Organization identifier
• temperature (float, optional): Sampling temperature (0.0-1.0)
• max_tokens (int, optional): Maximum tokens in response
• model (str, optional): Model to use for this request

Returns:

• dict: Response containing output text, token counts, and metadata

get_usage_stats(**filters)

Get usage statistics with optional filtering.

Parameters:

• customer_id (int, optional): Filter by customer
• organization_id (int, optional): Filter by organization
• start_date (datetime, optional): Start date for filtering
• end_date (datetime, optional): End date for filtering

Returns:

• dict: Usage statistics including request counts, token usage, and performance metrics

close()

Close database connections and clean up resources.

Contributing

We welcome contributions! Please see our Contributing Guide for details.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Changelog

See CHANGELOG.md for a list of changes and version history.

Acknowledgments

• Thanks to all contributors who have helped shape this project
• Built with love for the AI/ML community

---