Local-first vector memory with a 4-layer associative graph. Your agent remembers why things are connected, when things changed, and which memories matter most.
Conversation turns, tool outputs, observations. Any unstructured agent context fed in as text.
Every memory is evaluated: ADD new info, UPDATE existing, DELETE contradictions, or NO_OP if already known. Zero duplicates.
Persisted across 4 graph types in SQLite. Survives all session resets. REM cycle compresses while idle.
Similarity between memories. Finds related concepts across your full context history.
Cause → Effect relationships. Understands why things happened, not just what.
Before → After sequences. Tracks how knowledge evolves and decays over time.
Named entity co-occurrence. Connects people, projects, and events automatically.
Most vector stores are passive. They store what you put in and return what you ask for. VEKTOR is an active memory layer — it evolves, curates, and reasons about what your agent should remember.
Standard RAG vector stores
VEKTOR Memory
Drop-in memory layer for LangChain agents. recall() returns context, remember() stores every turn. v1 + v2 adapters included.
Wrap your OpenAI agent loop with persistent memory. Inject recalled context directly into system prompt. GPT-4o and o-series models supported.
Full MCP server module — vektor_recall, vektor_store, vektor_graph, vektor_delta tools. Connect Claude Desktop to persistent memory in minutes.
Provider-agnostic. Pass gemini, openai, groq, or ollama as provider. Key pooling for Gemini — waterfall rotation across up to 9 API keys, zero rate-limit downtime.
vektor_memoire HTTP tool — works with Le Chat and Mistral API agents. Local bridge on localhost:3847. Run mistral-setup.js to activate. French-first sovereign memory.
// 1. Install
// npm install vektor-slipstream
import { createMemory } from 'vektor-slipstream';
// 2. Initialise
const memory = await createMemory({
provider: 'gemini',
apiKey: process.env.GEMINI_API_KEY,
agentId: 'my-agent',
dbPath: './my-agent.db',
});
// 3. Remember — AUDN decides ADD/UPDATE/DELETE
await memory.remember("User prefers TypeScript");
// 4. Recall
const ctx = await memory.recall("coding preferences");
// 5. Traverse the graph
const g = await memory.graph("TypeScript", { hops: 2 });
// 6. What changed in 7 days?
const d = await memory.delta("architecture", 7);Pure SQLite. No cloud dependency, no API keys for memory. Your memory graph never leaves your server. LLM providers process queries per their own privacy policies.
Gemini, OpenAI, Groq, Ollama. Switch provider with one config change. Key pooling for Gemini — waterfall rotation across up to 9 keys.
Automatic curation loop prevents contradictions and duplicates. The graph stays consistent without any manual management.
Background process compresses 50 fragments into 3 core insights. Runs while your agent is idle. node rem.js or trigger via memory.dream().
The graph type system underpinning VEKTOR's four memory layers — semantic, causal, temporal, and entity.
GRAPH ARCHITECTURE ARXIV // 2601.02163Self-organizing memory architecture for structured long-horizon reasoning. Informs VEKTOR's REM cycle approach to memory consolidation and lifecycle management across extended agent sessions.
MEMORY LIFECYCLE ARXIV // 2504.19413Scalable long-term memory via dynamic extraction, consolidation, and graph-based retrieval. Informs VEKTOR's AUDN curation loop and REM compression cycle. Mem0 benchmarks show 90%+ token cost reduction — consistent with VEKTOR's synthesis approach.
MEMORY COMPRESSION LETTA.COMThe memory-as-OS paradigm that inspired VEKTOR's approach to agent context management and recall.
AGENT OSZero-knowledge architecture · Sovereign identity · Slipstream · Cloak · Pricing