Aethene is an open-source AI memory layer designed to enhance the memory capabilities of AI applications. This robust solution allows for effective storage, searching, and recalling of contextual information, thus providing an intelligent memory solution for AI systems.

Why Aethene?

Creating AI applications that possess memory functionality presents multiple challenges, including:

• Extracting important information from interactions
• Managing contradictions and updates effectively
• Conducting semantic searches across extensive memory databases
• Keeping track of changes through versioning
• Scaling efficiently while managing costs

Aethene addresses these challenges with a single API that offers virtually unlimited memory capabilities.

Key Features

• 🧠 Intelligent Extraction: Automatically extracts facts, preferences, and events from raw data without the need for manual tagging.
• 🔍 Hybrid Search: Combines vector similarity with recency boosts and intent comprehension to effectively locate the required information.
• 📝 Memory Versioning: New versions are created with each update to track historical memory changes and detect contradictions automatically.
• 🏢 Multi-Tenant Ready: Utilizes container tags to segment data per user, project, or organization with scoped API keys ensuring fine-grained access control.
• 🕸️ Entity Graph: Automatically generates relationship graphs by extracting data about people, places, and organizations.
• ⚡ Blazing Fast: Built on Convex, providing real-time, serverless performance with a P95 latency of under 200ms.

Example: Interfacing with Aethene API

Here’s a sample implementation demonstrating how to store and recall memories through the Aethene API:

# Store a memory
curl -X POST https://api.aethene.com/v1/content \
  -H "Authorization: Bearer $API_KEY" \
  -d '{"content": "User loves hiking and lives in San Francisco"}'

# Recall the stored memory
curl -X POST https://api.aethene.com/v1/recall \
  -H "Authorization: Bearer $API_KEY" \
  -d '{"query": "outdoor activities the user enjoys"}'
# Expected Response: "User loves hiking" with relevant context.

Additional Information

Aethene's architecture is designed for scalability and efficiency, utilizing modern technologies and methodologies.

For developers looking to integrate Aethene into their projects, in-depth API documentation is provided, detailing all endpoints and their respective functionalities. Full documentation is accessible in the openapi.yaml file included in the repository.

Getting Involved

Contributions to Aethene are welcomed. Potential contributors are encouraged to review the contributing guidelines and participate in enhancing the project further. A test suite is available for validation and integration testing.

For more information, visit the Aethene GitHub repository or reach out via the project's Twitter handle.