MemWal (now Walrus Memory)

https://decrypt.co/365834/agentic-memory-walrus-takes-on-ais-next-big-bottleneck

Walrus is a Sui-blockchain decentralised storage protocol; MemWal is its AI-facing SDK that stores encrypted agent memory on-chain — portable, verifiable, shareable across agent systems with no single-vendor lock-in. Targets production agents needing auditability. Renamed Walrus Memory (memwal.ai redirects to memory.walrus.xyz); still in beta, SDK published as @mysten-incubation/memwal.

Renamed. now Walrus Memory Source · Alternatives and migration notes →

Renamed · Verified 2026-10-29 · fresh as of 2026-10-29 · 41 cited facts, 5 with a third-party source · 2 recorded changes · Suggest a correction

At a glance

Tier
T2
Created
2025-01 vendor fix
License
Apache-2.0 repo fix
Freshness
2026-10-29 fix
Layer
agent-infra repo fix
Licensing
open-source repo fix
Deployment locus
hybrid repo fix
Platform vs product
platform repo fix
SDK languages
typescript registry fix

Memory taxonomy

storage
proprietary
retrieval
exact-match
persistence
lifelong
update
append-only
unit
document
governance
auditable
conflict
append

When to use

Optimised for: verifiability + portability + no vendor lock-in (on-chain)

Anti-fit: not for low-latency / high-write-throughput workloads (chain settlement)

Pros & cons

Pros

Walrus decentralized-storage backed memory — first memory layer with on-chain durability.

Cons

Web3-native — adds crypto / decentralization complexity that most teams don't need.

Claims & capabilities

$140M private token sale ($WAL, 2025) at ~$2B protocol valuation. Plugins for OpenClaw and NemoClaw.

Technical surface

API surface
searched not found
Backend storage
searched not found
Deployment
Decentralised blockchain storage (Sui blockchain via Walrus protocol; SDK in beta)
Embedding model
searched not found
Multi-tenancy
searched not found
MCP
not documented publicly
A2A
not documented publicly
OpenTelemetry
not documented publicly

Similar systems

Other dedicated memory layers in the catalog, ranked by inbound references.

  • Mem0 T1

    Universal memory layer for AI agents. Three concurrent stores (vector + graph + KV); LLM-extracted facts; concurrent retrieval via ThreadPoolExecutor.

  • Zep & Graphiti T1

    Bi-temporal knowledge graph (event time + ingestion time). Strong on chronological reasoning and contradiction tracking. Graphiti is the open-source core.

  • Cognee T1

    "Extract–Cognify–Load" pipeline that turns raw input into a typed, queryable knowledge graph for agent recall.

  • Hindsight (Vectorize) T1

    Standalone memory service from Vectorize. Open source. Biomimetic four-network design (World, Bank, Observation, Opinion). Ships an MCP memory server.

  • Letta / MemGPT T1

    Originated as MemGPT (UC Berkeley), where the model pages its own memory between context, recall and archival tiers. The product now centres on Letta Code, a stateful agent harness whose agents rewrite their own memory, skills and prompts over time.

  • Memori T2

    Structured, persistent memory layer for AI agents. Open source.

Sources (40 cited facts, 5 third-party)

Each fact links to the page it came from; the badge says who published it (how to read sources).

What changed

Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)

  1. 2026-10-29
    • Name “MemWal (Walrus)” → “MemWal (now Walrus Memory)”
    #305
  2. 2026-10-22
    • License “not applicable — not OSS” → “Apache-2.0” source
    • Latest release “not applicable — not OSS” → “@mysten-incubation/memwal 0.1.5 (Aug 26; page shows no year, presumably 2026)” source
    #195