Vic563 Memgpt
CreatedMCP-Mirror
Implements a memory system for large language models, enabling conversational context persistence across multiple LLM providers.
About
This TypeScript-based server provides a memory system for LLMs, allowing users to chat with various LLM providers (OpenAI, Anthropic, OpenRouter, Ollama) while maintaining conversation history. It offers tools to send messages, retrieve past conversations, clear memory, and switch between providers and models. The server supports the Claude 3 and 3.5 series models, as well as OpenAI's GPT models, and allows for unlimited memory retrieval.
Key Features
- Supports unlimited memory retrieval.
- 0 GitHub stars
- Supports multiple LLM providers including OpenAI, Anthropic, OpenRouter, and Ollama.
- Offers tools to send messages, retrieve conversation history, and clear memory.
- Maintains conversation history with memory management features.
- Allows switching between different models for each provider.
Use Cases
- Building conversational AI applications with persistent memory.
- Integrating memory capabilities into existing LLM workflows.
- Experimenting with different LLM providers and models within a unified memory system.
