HN Showcase
Back to browse
GitHub Repository

Verbatim semantic memory across Claude Code sessions - pgvector + local Ollama embeddings

3 starsPython

81% on LongMemEval with Pgvector and BM25 – Memory for Claude Code

by m4cd4r4·Apr 9, 2026·2 points·2 comments
Visit ProjectView on HN

AI Analysis

●●SolidBig BrainSolve My Problem

Pgvector + BM25 hybrid retrieval with honest LongMemEval benchmarks for Claude Code memory.

Strengths
  • •81% LongMemEval score with full per-category breakdown and reproduction steps included.
  • •Native Claude Code hook integration via single-file JavaScript, ~10 minute install.
  • •Verbatim storage only—no LLM summarization deciding what's important to keep.
Weaknesses
  • •Only useful if you live in Claude Code; other assistants need different solutions.
  • •Memory tools for AI assistants already exist (MemPalace, mem0, letta).
Target Audience

Developers using Claude Code who lose context between sessions

Similar To

MemPalace · mem0 · letta

Similar Projects

Developer Tools●Mid

CogmemAi – Persistent Memory for Claude Code via MCP

Runs extraction and search server-side so your local MCP is a tiny HTTP client — no local DBs, no giant RAM leaks, and an easy npx install and .mcp.json or global MCP registration. It exposes clear tools (save_memory, recall_memories, extract_memories, get_project_context) and adds project-scoped + global preferences — a pragmatic fix for Claude Code's tiny flat-file memory. The tradeoff is obvious: usefulness depends on the hosted API (privacy, uptime, cost), and the repo looks early-stage with minimal commits and docs beyond the quickstart.

Niche GemShip ItSolve My Problem
hifriendbot
203mo ago
Developer Tools●●Solid

Fixing Claude Code's amnesia with persistent memory

Scatter-gather retrieval with prompt caching beats naive vector search for Claude Code memory.

Big BrainSolve My Problem
NBenkovich
132mo ago