AI/ML●●Solid
Engram – open-source persistent memory for AI agents (Bun and SQLite)
SQLite-backed agent memory with graph viz when Mem0 and Zep already dominate.
SlickShip It
zanfiel
333mo ago
Back to browse
Vektor – local-first associative memory for AI agents
by vektormemory·Apr 5, 2026·2 points·0 comments
AI Analysis
●●SolidSlickBig Brain
Local-first agent memory with SQLite graphs, but requires a license key.
Strengths
- •Bundled ONNX models mean zero external API dependencies for embeddings.
- •SQLite graph storage ensures portable, single-file persistent memory without servers.
- •Claude Desktop .dxt extension offers drag-and-drop integration without config files.
Weaknesses
- •Commercial license required for production, limiting community adoption and testing.
- •Cloak stealth browser feature feels disjointed from the core memory product.
Category
Target Audience
AI agent developers, Privacy-focused engineers
Similar To
Mem0 · Zep · LangChain Memory
Post Description
4-layer memory via MAGMA graph in SQLite. AUDN curation loop (add/update/delete/no-op). REM background compression. Claude tools. No cloud. npm i vektor-slipstream
If anyone has deep experience in DB product testing and wants to test out the final release 1.3.6, reach out via the emal/website and mention you read this via HN thread and we will provide a free product test code license. We do need detailed feedback: [email protected]
Similar Projects
Developer Tools●●Solid
Local semantic memory for coding agents
Local-only agent memory when mem0 and supermemory require servers.
CozyNiche Gem
chadiiek
201mo ago
Developer Tools●●●Banger
Agent Recall – Open-source, local memory for AI agents (SQLite/MCP)
Scope-hierarchical memory for agents across projects—Mem0 and Zep flatten context too much.
Solve My ProblemBig Brain
MaxNardit
123mo ago
AI/ML●●●Banger
Clawlet – AI agent with built-in semantic memory, one binary
Single binary with zero dependencies ships vector semantic memory in SQLite.
Ship ItDark HorseWizardry
mosaxiv
514mo ago
AI/ML●●Solid
AFS – filesystem-native memory layer for AI agents
Putting the memory layer on-disk as a .afs/ tree is a gutsy, practical move — you get searchable JSON, FTS5 for text queries, HNSW vectors for similarity, and msgpack edges for relationships without running a separate DB service. It feels like a thoughtful toolkit for agents that must persist thinking artifacts (observations → reflections → knowledge), though I want to see details on concurrency, index portability, and how this performs at scale before betting production workloads on it.
Big BrainBold Bet
thompson0012
203mo ago
AI/ML●●Solid
AgentKV – SQLite for AI agent memory (MMAP vector+graph DB)
Single-file mmap storage plus an HNSW vector index and explicit graph edges is an elegant, practical combo — think "SQLite for agent memory" with CRC-32 crash recovery and zero-server convenience. The C++20 core + nanobind gives zero-copy NumPy views and GIL-free searches, and the claimed FAISS-like throughput makes this genuinely interesting for local setups; main caveat is build/toolchain friction and how rich the surrounding ecosystem becomes.
WizardryNiche Gem
shiwang_khera
104mo ago