Motorhead is a deprecated memory and information retrieval server designed to simplify state management for LLM-powered chat applications. It offers three core APIs for managing conversation history, enabling incremental summarization, and performing text-based retrieval using Vector Similarity Search (VSS).

How It Works

Motorhead stores conversation messages in sessions, automatically summarizing older messages to manage context window limits. When the message window reaches a configured maximum (MOTORHEAD_MAX_WINDOW_SIZE), it triggers a summarization process, reducing the token count and maintaining a concise history. For long-term memory, it leverages Redisearch VSS, allowing efficient retrieval of relevant information based on text queries, segmented by session ID.

Quick Start & Requirements

• Install/Run: docker-compose build && docker-compose up or docker pull ghcr.io/getmetal/motorhead:latest and run.
• Prerequisites: Redis instance (REDIS_URL environment variable required), OpenAI API key (OPENAI_API_KEY). Azure deployments require additional Azure-specific environment variables.
• Configuration: MOTORHEAD_MAX_WINDOW_SIZE (default 12), MOTORHEAD_LONG_TERM_MEMORY (default false), MOTORHEAD_MODEL (default gpt-3.5-turbo), PORT (default 8000), OPENAI_API_BASE (default OpenAI v1).
• Links: Chat JS Example, Vanilla JS Example, Hosted Example, Python Example

Highlighted Details

• Provides APIs for GET, POST, and DELETE operations on session memory.
• Supports incremental summarization of conversation history.
• Integrates with Redisearch for long-term memory and VSS retrieval.
• Configurable model for summarization (gpt-3.5-turbo or gpt-4 recommended).

Maintenance & Community

The project is marked as DEPRECATED and support is no longer maintained.

Licensing & Compatibility

The license is not specified in the README.

Limitations & Caveats

The project is explicitly marked as DEPRECATED, meaning it will not receive further updates or support, posing a significant risk for adoption. The README does not specify the license, which could impact commercial use or integration into closed-source projects.