MemoryAtlas

MemoryBear

RedBear AI (Suanmo Suanyang Technology)

Next-generation AI memory system inspired by hippocampal memory encoding and neocortical knowledge consolidation. Spans the full knowledge lifecycle: perception → extraction → association → forgetting. LLM-driven extraction converts conversations into structured entity-relationship triples stored in a Neo4j knowledge graph, while a parallel vector store enables hybrid semantic+keyword retrieval. A biologically-motivated forgetting engine (dormancy → decay → clearance) prunes low-value knowledge automatically.

Storage
Dual-store architecture: Neo4j knowledge graph stores extracted entity-relationship triples with temporal anchors and provenance; a vector store (BERT embeddings) holds parallel semantic representations for similarity search. LLM extraction at write time converts raw text into structured triples and syncs them to both stores. Each knowledge item carries a dynamic memory-strength score updated by usage frequency and association activity.
Retrieval
Dual-engine hybrid retrieval: BERT-based semantic vector search (synonym and implicit-intent aware) expands the candidate space; BM25 keyword search then performs precise filtering over candidates. The two signals are fused via RRF. Interactive knowledge-graph visualisation available for human inspection. Self-reported internal benchmark: 92% retrieval accuracy vs 57% single-mode baseline (vendor measurement, untracked benchmark — config not published).
Self-host
Self-host: heavy
License
Apache-2.0
Pricing
Open-source Apache-2.0. Self-hosting requires Neo4j (non-trivial to operate). The commercial cloud at memorybear.ai now publishes tiers (page is Chinese-language, prices quoted in USD): Free Experience $0/mo (1 space, 5 skills, 2 apps, 1 knowledge base at 300MB, 1 memory engine, 10 end users, community support); Personal Productivity $19/mo (20 skills, 10 apps, 3 KBs at 1GB each, 3 standard memory engines, 50 end users); Professional $299/mo (3 spaces, unlimited skills, 30 apps, 10 KBs at 10GB each, 10 advanced memory engines, 500 end users, ticket support); Business OEM is on-premises with pricing on request (20,000 memory records, 10,000 smart searches/mo, 99.9% SLA). · Free + paid
GitHub stars
5,034
Last release
2026-07-29
Last commit
2026-08-07
First catalogued
2026-06-28

Strengths

  • Full knowledge lifecycle with biologically-motivated forgetting: dormancy → decay → clearance keeps memory collections pruned rather than accumulating stale data
  • Dual Neo4j + vector store architecture supports rich relational queries (graph traversal) alongside semantic similarity — stronger than pure vector-only memory
  • Multimodal affective intelligence path (arXiv:2603.22306) extends to audio, image, and emotion-aware memory for assistants
  • v0.3.9 (Jun 2026, codename PoJun) added event-timeline editing on entity descriptions (add/update/delete individual events, not just whole-fact overwrite), a latency-optimized Express Retrieval mode, and multi-relationship entity types
  • Memory MCP service listed on the ModelScope MCP marketplace as of v0.3.9 — a distribution channel distinct from a first-party MCP server entry in this catalog until independently verified
  • v0.3.12 (Jul 2026, codename MingJing) made Express Retrieval and Persona Memory first-class Agent-callable tools and added activity-aware forgetting (memories are pruned by real creation time + usage frequency, not just static rules)
  • 4860 stars, actively maintained with frequent releases (roughly weekly since v0.3.9); Apache-2.0

Watch out

  • Neo4j dependency makes self-hosting heavier than most catalog entries — production deployments need a managed or self-hosted Neo4j instance
  • Organisation is a Chinese entity (Suanmo Suanyang Technology / RedBear AI); evaluate long-term maintainership and supply-chain considerations
  • Self-reported 92% retrieval accuracy is on an untracked internal benchmark with no published config — treat as directional, not comparable to LoCoMo/LongMemEval results
  • Cloud commercial tier (memorybear.ai) pricing not publicly listed — the marketing site is a JS SPA with no reachable pricing page as of 2026-07-23
  • v0.3.9 announced removal of the sync-write/async-read endpoints and deprecated the async status-check endpoints; v0.3.10 (2026-07-01) confirmed both removed outright — integrations still tracking the old API surface must migrate
  • 'Memory MCP' is referenced repeatedly in release notes (persistence in v0.3.10, native Agent tool-calling in v0.3.12) but no first-party MCP server path is discoverable in the open-source repo — treat as a hosted/cloud-product capability, not a self-hostable MCP integration, until a repo location is confirmed

Best for

  • AI assistants and agents that need rich relational knowledge management — especially where entity relationships, temporal tracing, and automated forgetting matter more than pure vector recall speed

Benchmark results

No sourced results yet.

Sources

Last verified 2026-08-01 · updated by refresh-framework-cards