Back to browse
GitHub Repository

Cognitive memory for AI agents — learns from use, forgets what's irrelevant, strengthens what matters. Single binary, fully offline.

214 starsRust

Shodh– AI memory that learns from use, no LLM calls, single Rust binary

by Varun_shodh·Feb 28, 2026·6 points·1 comment

AI Analysis

●●●BangerWizardryBig BrainZero to One

Offline AI memory that learns without LLM calls—Hebbian decay plus vectorization beats mem0 architecture.

Strengths
  • Genuinely novel approach: Hebbian learning + mathematical decay instead of API-dependent competitors
  • Zero external dependencies (no OpenAI, Neo4j, vector DB) with 55ms memory storage vs 20+ seconds for alternatives
  • Multi-language support (Rust, Node, Python, Docker) with MCP integration for Claude/Cursor
Weaknesses
  • Early ecosystem (104 GitHub stars, minimal production case studies)
  • Comparison table vs mem0/Cognee/Zep claims superiority but lacks independent benchmarks
Target Audience

AI agent developers, self-hosted AI enthusiasts

Similar To

mem0 · Cognee · Zep

Similar Projects

AI/MLMid

Ebbforge - 10M agent Rust swarm engine, 8 fundamental benchmarks

Rust swarm vs LLM agents is clever positioning, but benchmarks are self-designed and lack third-party validation.

Big BrainWizardry
agent-world
213mo ago
Developer Tools●●●Banger

MuninnDB – ACT-R decay and Hebbian memory for AI agents

ACT-R decay and Hebbian learning as native primitives, not vector hacks.

Big BrainWizardryNiche Gem
mjbonanno
273mo ago