by jcartu · MCP Server · ★ 32
RASPUTIN Memory — Experimental Conversational Memory System A research codebase exploring fact-type partitioned retrieval for LLM agent memory. Not state of the art. Published for the architectural insights. Status: Archived This project is archived and no longer maintained. It is published as a public research artifact at the final state of (commit , 2026-04-25). No issues will be triaged, no pull requests reviewed, no further development undertaken. Forks are welcome under the MIT license. The closure rationale, final benchmark, key insights, and pointers for future work are documented in .
| Stars | 32 |
| Forks | 5 |
| Language | Python |
| Category | MCP Server |
| License | MIT |
| Quality Score | 62.4401967314605/100 |
| Open Issues | 19 |
| Last Updated | 2026-04-25 |
| Created | 2026-03-26 |
| Platforms | claude-code, mcp, python |
| Est. Tokens | ~4829k |
Explore other popular mcp server tools:
rasputin-memory is The ultimate memory backend for OpenClaw and Claude Code. Persistent conversation memory with LLM fact extraction, foundation-model reranking, and 77.7% LoCoMo accuracy. MCP native, REST API, self-hos. It is categorized as a MCP Server with 32 GitHub stars.
rasputin-memory is primarily written in Python. It covers topics such as agent-memory, ai, ai-memory.
You can find installation instructions and usage details in the rasputin-memory GitHub repository at github.com/jcartu/rasputin-memory. The project has 32 stars and 5 forks, indicating an active community.
rasputin-memory is released under the MIT license, making it free to use and modify according to the license terms.