Self-hosted long-term memory substrate for any LLM-using tool. Hybrid vector + lexical recall with MMR reranking, auto-extracted knowledge graph, sleep-style consolidation, per-project tuning, MCP server, REST + OAuth 2.0, embedded React admin UI. SQLite or Postgres + pgvector or Qdrant. Pluggable providers (OpenAI, Anthropic, Gemini, Ollama).

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "nram": {
      "command": "go",
      "args": [
        "run",
        "github.com/nram-ai/nram@latest"
      ]
    }
  }
}

README Excerpt

> **Work in Progress:** This project is under active development. Some features may be incomplete, unstable, or subject to change. Contributions and feedback are welcome, but expect rough edges as things continue to evolve. A self-hosted, long-term memory layer that any LLM-using tool can plug into. Store, recall, and enrich memories with vector embeddings, knowledge graphs, sleep-style consolidation, and a built-in admin UI.

Tools (20)

DATABASE_URLDELETEGETLOG_LEVELNRAM_ADMIN_EMAILNRAM_ADMIN_PASSNRAM_CONFIGPORTPOSTPUTmemory_delete_projectmemory_enrichmemory_exportmemory_forgetmemory_getmemory_graphmemory_listmemory_projectsmemory_recallmemory_store

Topics

ai-agentsai-memorychatgptclaudecontext-engineeringembeddingshybrid-searchknowledge-graphllm-toolslong-term-memorymcp-servermemory-consolidationmulti-tenantoauth2query-expansion

nram

Quick Install

README Excerpt

Tools (20)

Topics

Related AI/LLM Servers