MemPalace
Spatial-metaphor memory (wings / rooms / halls) on top of verbatim chunked storage. Independent analysis suggests the score is driven by verbatim storage + ChromaDB defaults rather than the palace structure.
Active · Verified 2026-10-28 · fresh as of 2026-10-28 · 28 cited facts, 1 with a third-party source · Suggest a correction
At a glance
- Type
- Verbatim chunks + spatial metaphor vendor fix
- Tier
- T2
- Section
- Research / specialised systems
- Latest release
- v3.3.4 2026-05-01 repo fix
- GitHub
- 51.3k★ +49.6k/mo Python repo fix
- Pricing
- searched not found fix
- Funding
- not applicable — not commercial fix
- Freshness
- 2026-10-28 fix
- Layer
- agent-infra repo fix
- Licensing
- open-source repo fix
- Developer activity
- OpenRank 16.11 (12-mo avg 24.34) · 106 participants — 2026-09 third-party fix
Memory taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- long-term
- update
- append-only
- unit
- chunk
- governance
- inspectable
- conflict
- append
When to use
Optimised for: research positioning (see memory_model)
Anti-fit: not applicable - research paper
Pros & cons
Pros
Three-layer corpus model (raw / governed / substrate) with explicit governance phases — most architecturally principled OSS memory framework.
Cons
OSS framework; less polished than commercial memory layers; smaller user base.
Claims & capabilities
Reports 96.6% Recall@5 on LongMemEval (later updated to 98.4% on held-out data after benchmark-tuning controversy).
Technical surface
- API surface
- searched not found
- Backend storage
- searched not found
- Deployment
- searched not found
- Embedding model
- searched not found
- Multi-tenancy
- searched not found
- MCP
- no first-party MCP adapter published as of 2026-05; community connectors may exist.
- A2A
- no Google A2A (Agent2Agent) integration documented as of 2026-05.
- OpenTelemetry
- no first-party OpenTelemetry exporter documented; standard logs/metrics typically available.
Similar systems
Other research / specialised systems in the catalog, ranked by inbound references.
- A-MEM T3
Treats memories as atomic linkable notes — explicit nod to Zettelkasten knowledge management. Dynamic linking; retroactive memory revision.
- BAI-LAB MemoryOS T3
Hierarchical "OS" with Storage / Updating / Retrieval / Generation modules. Short-term → mid-term via FIFO dialogue-chain; mid-term → long-term via segmented paging.
- Titans (Google) T4
Neural long-term memory module that learns to memorise at test time. Uses gradient-of-loss as "surprise" signal; adaptive weight-decay forgetting. Three variants: MAC (memory-as-context), MAG (memory-as-gate), MAL (memory-as-layer).
- EverMemOS T4
Self-organizing memory OS for structured long-horizon reasoning. Three-phase model: episodic, semantic, reconstructive.
- EVOLVE-MEM T3
Dynamic Memory Network + Hierarchical Memory Manager + Self-Improvement Engine. L0 raw embeddings, L1 contextual summaries, L2 high-level principles. NeurIPS 2025 (Scaling Environments for Agents workshop).
- LiCoMemory T4
Lightweight hierarchical graph (CogniGraph) with entities and relations as semantic indexing layers. Incremental graph construction, fast updates, low-latency inference. Nov 2025.
Sources (27 cited facts, 1 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Verbatim chunks + spatial metaphor vendor
- What it does & what's distinct Spatial-metaphor memory (wings / rooms / halls) on top of verbatim chunked storage. Inde… vendor
- Claims / benefits Reports 96.6% Recall@5 on LongMemEval (later updated to 98.4% on held-out data after ben… vendor
- Modalities text vendor
- Created 2026-04 vendor
- Latest release v3.3.4 2026-05-01 repo
- License MIT repo
- GitHub 51.3k★ +49.6k/mo Python repo
- Founders / pedigree no data — searched mempalace.net; no author or affiliation listed vendor
- Memory volume / scale no scale claim — research / open-source vendor
- Performance 96.6% LMES (vendor-self-reported; flagged for verification) vendor
- Versioning immutable append-log vendor
- Tombstoning append-only no delete vendor
- Schema evolution schema with migration vendor
- Namespace primitives per-tenant vendor
- Contradiction handling always append (no resolution) vendor
- Forgetting policy none (permanent) vendor
- Import / export JSON via REST API (vendor docs). vendor
- Validated verticals research only vendor
- Optimised for research positioning (see memory_model) vendor
- Pros Three-layer corpus model (raw / governed / substrate) with explicit governance phases — … vendor
- Cons OSS framework; less polished than commercial memory layers; smaller user base. vendor
- Project & sources GitHub MemPalace/mempalace mempalace.net arxiv 2604.21284 (critical analysis) Vectorize … vendor
- commit-trajectory [{"ym":"2026-04","cum":691},{"ym":"2026-05","cum":927}] repo
- licensing-bucket open-source repo
- taxonomy-three-layer-rollup agent-infra repo
- developer-activity-opendigger {"activity":239.01,"as_of":"2026-09","openrank_12mo":24.34,"openrank_current":16.11,"par… third-party
Related systems
References (1)
- Chroma builds on — Independent analysis suggests the score is driven by verbatim storage + ChromaDB defaults rather than the palace structure.