Zero-LLM graph memory
for every AI agent.
Free to start.
Zero LLM inference at write time — no token surprises, no write-time LLM call. $0 ingest cost, sub-millisecond in-process engine ops, graph-native from day one — powered by GraphPalace's Rust/LadybugDB engine. 35 MCP tools. Works with anything.
No credit card required · 3 agents · Graph memory included
Zero LLM calls · LadybugDB graph engine · Pheromone-weighted retrieval
Graph memory in minutes, not months.
Five lines to connect. One call to write. Zero config to get graph-native entity extraction. No embedding pipelines, no LLM chains, no separate graph database to manage.
- ✓Connect via the MCP endpoint + cURL — no SDK install needed
- ✓Connect any agent framework in < 5 minutes
- ✓35 MCP tools — Claude Desktop, Cursor, LangChain, CrewAI, FastAPI, Google ADK
- ✓Multi-hop graph queries out of the box
import { AgentverseMemory } from "@fetchai/agentverse-memory"; const memory = new AgentverseMemory({ apiKey: process.env.AM_API_KEY,}); // Store — no LLM at write; sub-ms in-process engine opawait memory.store({ content: "Alice prefers dark roast coffee", type: "semantic",}); // Search — hybrid retrieval (TF-IDF + dense, RRF)const result = await memory.search( "what does Alice prefer?");// → { results: [...], retrieval: "hybrid" }Built different. Priced differently.
GraphPalace's Rust/LadybugDB architecture changes what's possible at each price point. When writes don't require LLM calls, graph memory can be free.
Zero-LLM Writes
No LLM calls at write time. GraphPalace uses TF-IDF + LadybugDB — the in-process engine write op is sub-millisecond (0.035ms p95 in am-local), so writes never wait on, or bill for, an LLM-extraction call the way LLM-based systems do. Predictable cost, no token surprises — $0 ingest, zero-LLM writes.
Graph Memory at Every Tier
Knowledge graphs, entity relationships, and multi-hop queries — included on the free tier. Mem0 gates graph at $249/mo; RetainDB at $9/mo. We give it away free, with zero-LLM writes.
Native MCP Integration
35 Model Context Protocol tools over JSON-RPC 2.0. Drop into Claude Desktop, Cursor, LangChain, CrewAI, FastAPI, Google ADK — any MCP-compatible framework. Published as io.github.fetchai/agentverse-memory on the MCP Registry.
Pheromone-Guided Retrieval
Stigmergic memory trails with 5 decay types. The most relevant memories surface naturally — no manual ranking, no prompt engineering.
4 Memory Types
Episodic, semantic, procedural, and working memory — all four types available at every tier. Each backed by the LadybugDB graph engine (active fork of Kùzu).
Cross-Agent Memory Sharing
5 shared space tools with JWT/DID-authenticated access. Create, join, store, and query across agents — securely, verifiably. Self-host with Docker Compose or use managed Cloud Run.
How we compare
Mem0, Zep, and Google Memory Bank all invoke an LLM on every write. Agentverse Memory never does — that's the moat: $0 ingest cost, no write-time token bill, and graph memory free at every tier.
| Feature | Agentverse Memory Free → $99/mo | Mem0 $19 → $249/mo | RetainDB $0 → $99/mo | Zep $125/mo+ |
|---|---|---|---|---|
| LLM call at write time | Never | Every write | Extraction step | Every write |
| Ingest cost (LLM tokens) | $0 ingest | Per-write tokens | Per-write tokens | Per-write tokens |
| Engine write op (in-process) | 0.035ms p95 | Not published | Not published | Not published |
| Pheromone retrieval | ✓ | ✗ | ✗ | ✗ |
| Active Inference agents | Pro+ | ✗ | ✗ | ✗ |
| MCP tools | 35 tools | Limited | MCP server | Limited |
| Graph included on free tier | ✓ | ✗ | ✗ | ✓ |
| Lowest tier with graph | Free | $249/mo | $9/mo | Free |
Competitor pricing verified June 2026 (mem0.ai, getzep.com, retaindb.com). Mem0, Zep, and Google Memory Bank perform LLM-based extraction at write time; RetainDB performs extraction via its capture step (per-write LLM use undocumented). Zep includes its graph on the free tier; Mem0 gates graph at $249/mo (Pro) and RetainDB at $9/mo (Builder). "$0 ingest" reflects Agentverse Memory's zero-LLM write path — measured at $0 vs Mem0 ~$0.063 / Graphiti ~$0.36 per 500-query ingest in our reproducible head-to-head. Engine write p95 0.035ms is an in-process (am-local) micro-benchmark, not deployed end-to-end; deployed writes are ~150ms. Mem0's published ~2.59s is a p95 query latency (graph variant), not a write latency.
RetainDB's graph starts at $9/mo — and there's no graph on its free tier.
RetainDB populates its knowledge graph using LLM-based extraction in its capture step (per-write LLM use is undocumented), so ingest carries token costs that scale with every byte you store — and its graph isn't available on the free Starter tier (it begins at $9/mo Builder).
Agentverse Memory uses TF-IDF + LadybugDB with zero LLM at write time — $0 ingest cost and the graph is included on every tier, free included. Writes never wait on, or bill for, an LLM call.
Why GraphPalace is different
Most memory systems call an LLM to extract entities at write time. GraphPalace uses TF-IDF + LadybugDB — classical information retrieval in a Rust binary. No API call, no token cost, no latency.
*Engine write p95 0.035ms is an in-process (am-local) micro-benchmark; deployed writes are ~150ms.
Simple pricing. Graph included.
No credit math. No graph gates. No token surprises.
Zero-LLM writes. Graph at every tier.
Start building memory-enabled agents today. Three agents, full graph support, pheromone retrieval — free, no credit card. No LLM at write time, zero token surprises.