Back to browse
GitHub Repository

Biologically-inspired memory for AI agents. Decay, retrieval strengthening, consolidation. Zero dependencies.

684 starsTypeScript

Hippo, biologically inspired memory for AI agents

by kitfunso·Apr 6, 2026·128 points·29 comments

AI Analysis

●●●BangerBig BrainZero to OneNiche Gem

Biological decay mechanics beat vector search for agent memory that actually forgets.

Strengths
  • Memory decay and strengthening model mimics human forgetting instead of storing everything
  • Portable markdown storage works across Claude Code, Cursor, and Codex without vendor lock-in
  • Explainable recall shows which terms matched and whether BM25 or embeddings contributed
Weaknesses
  • Requires Node.js 22.5+ which excludes some enterprise environments still on LTS
  • Optional embeddings need @xenova/transformers adding ~500MB for semantic search
Category
Target Audience

AI agent developers, multi-tool AI users (Claude Code, Cursor, Codex)

Similar To

Mem0 · LangChain Memory · Zep

Similar Projects

AI/ML●●Solid

AI memory with biological decay (52% recall)

Ebbinghaus forgetting curve cuts token waste 84% while doubling recall versus Zep.

Big BrainShip It
SachitRafa
98531mo ago
AI/ML●●●Banger

YantrikDB – persistent memory for AI agents

Bundled 7MB embedder means zero network calls or model downloads for agent memory.

Zero to OneWizardrySolve My Problem
pranabsarkar
111mo ago