Mpalo SDKs

Official SDKs for integrating Mpalo's AI memory capabilities into your applications. Our SDKs provide a developer-friendly interface to interact with Palo models and memory enhancement features.

Python SDK

The official Python SDK provides a comprehensive interface for interacting with Mpalo's API. Built with developer experience in mind, it offers intuitive methods, clear documentation, and robust error handling.

Developer-Friendly

Intuitive API design with clear naming conventions and comprehensive documentation.

Secure & Robust

Built-in security features, error handling, and retry mechanisms for reliability.

High Performance

Optimized for performance with support for async operations and streaming.

Installation

Install the Python SDK using pip:

pip install mpalo

The SDK requires Python 3.7 or higher and includes all necessary dependencies.

Quickstart

Get started with the Python SDK in minutes:

from mpalo import Mpalo

# Initialize with your API key
mpalo_client = Mpalo(api_key="your_api_key")

# Process interaction
response = mpalo_client.process_interaction("Tell me about quantum computing")
print(response)

Features

The Python SDK provides a rich set of features for working with Mpalo's API:

Examples

Here are some common examples of using the Python SDK:

# Create a conversation session
session_id = mpalo_client.create_conversation("Research Discussion")

# Process interactions with context
response1 = mpalo_client.process_interaction(
    "Tell me about quantum computing",
    session_id=session_id
)

response2 = mpalo_client.process_interaction(
    "What are its practical applications?",
    session_id=session_id
)

# Configure memory parameters
response = mpalo_client.process_interaction(
    "What do you know about AI?",
    memory_traversal=True,
    memory_mapping=True,
    top_k=5,
    min_score=0.7
)

Development Roadmap

Our SDK development roadmap includes exciting upcoming features:

Privacy & Security

Mpalo is built with privacy and security as foundational principles. All user data is encrypted in transit and at rest (AES-256 for advanced models). We do not use your data for training unless you give explicit consent. You retain ownership of your input and output, and can request deletion or review at any time. Our API and SDKs are designed to comply with GDPR and other major privacy regulations.

• Data encrypted at rest and in transit
• No training on user data without explicit consent
• Options for data access, review, and deletion
• Privacy-by-design: user control over memory and personalization
• Legal/IP protection: reverse engineering, extraction, or replication of models is strictly prohibited

Supported Vector Stores & LLM Providers

The Mpalo SDK and API support integration with a wide range of vector stores and external LLM providers. This enables flexible, secure, and scalable memory and model management for your applications.

Note: Each provider may require specific connection parameters. See the SDK and API reference for details and examples.

Advanced Features

• Rate Limiting: Protect your application and users with built-in rate limiting. Default limits are set slightly above those of major LLM providers for quick integration.
• Anomaly Detection: Monitor API usage for suspicious patterns (e.g., extraction attempts, abnormal query rates) with built-in anomaly detection and alerting.
• Session Management: Create, manage, and track conversation sessions for personalized, context-aware AI experiences.
• Memory Template Connections: Easily connect and manage memory templates for different use cases and vector stores.
• Custom Filters & Parameters: Fine-tune memory retrieval and LLM calls with advanced filtering and parameterization.
• Performance Monitoring: Track latency, throughput, and error rates for your API and SDK usage.

Ethics & Compliance

Mpalo is committed to ethical AI development and compliance with global standards. Our principles include fairness, transparency, accountability, privacy, and beneficence. We reinvest profits into research and prioritize people over profit.

• Bias mitigation and fairness in all models
• Transparency and explainability (where possible)
• Accountability and robust safety testing
• Privacy and data protection by design
• Human oversight and control
• Compliance with GDPR and major privacy laws

Contribution & Support

Mpalo SDKs are open source and we welcome contributions from the community. You can report bugs, request features, or contribute code via our GitHub repositories. For support, visit our Support Hub or join the Community Forum.

• Report bugs and issues on GitHub
• Request features and improvements
• Submit pull requests (see CONTRIBUTING.md)
• Contact support for help with integration or troubleshooting

Why Use the Mpalo SDK?

• Developer-friendly interface for rapid integration with Palo memory models.
• Advanced error handling, session management, and security features built-in.
• Optimized for performance and scalable for production workloads.
• Consistent API for all supported memory engines and features.
• Open source, extensible, and actively maintained.

Supported Languages

Python

PyPI GitHub

JavaScript

Coming Soon

Key Features

Resources

• SDK GitHub Repository
• PyPI Package
• API Reference
• Tutorials
• API Status