{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreigoh6myjkudsrak7bntqz4byfwiylwgnxddgetyf4w5rmht27cy2e",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mh2ycl3znud2"
  },
  "path": "/t/widemem-open-source-memory-layer-for-llms-with-importance-scoring-decay-and-conflict-resolution/174269#post_1",
  "publishedAt": "2026-03-15T01:13:38.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "GitHub - remete618/widemem-ai: Next-gen AI memory layer with importance scoring, temporal decay, hierarchical memory, and YMYL prioritization · GitHub",
    "widemem-ai · PyPI",
    "https://widemem.ai"
  ],
  "textContent": "Hi everyone, sharing an open-source project I’ve been working on.\n\n**widemem** is a Python memory layer for LLMs that goes beyond basic vector search. Most memory systems treat all facts equally and let contradictions accumulate silently. widemem tries to fix that.\n\n## What it does\n\n  * **Importance scoring** : each extracted fact gets a 1-10 score. Retrieval ranks by a weighted mix of similarity, importance, and recency\n  * **Time decay** : configurable exponential/linear/step decay. Old trivia fades, critical facts stick\n  * **Batch conflict resolution** : “I moved to Paris” after “I live in Berlin” gets resolved in a single LLM call instead of storing both\n  * **YMYL prioritization** : health, legal, and financial facts get higher importance floors and decay immunity\n  * **Hierarchical memory** : facts roll up into summaries and themes with automatic query routing\n  * **MCP server** : works as a Model Context Protocol server for Claude Desktop and other MCP clients\n\n\n\n## Runs fully local\n\nWorks with Ollama + sentence-transformers + SQLite + FAISS out of the box. No cloud, no API keys needed. Also supports OpenAI, Anthropic, and Qdrant if you want.\n\n\n    pip install widemem-ai\n\n\n140 tests passing. Apache 2.0.\n\n  * GitHub: GitHub - remete618/widemem-ai: Next-gen AI memory layer with importance scoring, temporal decay, hierarchical memory, and YMYL prioritization · GitHub\n  * PyPI: widemem-ai · PyPI\n  * Site: https://widemem.ai\n\n\n\nWould love feedback, especially from anyone working on agent memory or RAG pipelines. What approaches are you using for memory in your projects?",
  "title": "widemem: open-source memory layer for LLMs with importance scoring, decay, and conflict resolution"
}