MemoraX AI
Shenzhen. Argues attaching external vector stores is architecturally flawed; trains memory as an intrinsic capability via Agentic RL. Plans B2B enterprise KM and C2C personalisation in 12 months. Founded by Tianjin University AI professor. Company site memoraxai.com (operated by 深圳忆纪元科技有限公司) describes textual, work/coding and multimodal memory infrastructure; ships MemoraX Code, an MIT-licensed cross-agent coding memory for Codex and Claude Code (npm @memorax/memorax-code, first published 2026-08), plus public evals LoCoMo-Refined and ScriptMem.
Active · Verified 2026-10-29 · fresh as of 2026-10-29 · 25 cited facts, 17 with a third-party source · 1 recorded change · Suggest a correction
At a glance
- Type
- Endogenous / in-model memory (no external bank) third-party fix
- Tier
- T2
- Section
- Dedicated memory layers
- Latest release
- MemoraX Code 0.1.21 (npm @memorax/memorax-code, 2026-10-21) registry fix
- GitHub
- memorax-ai/memorax-code (MemoraX Code, MIT); org github.com/memorax-ai also hosts ScriptMem repo fix
- Pricing
- searched not found fix
- Funding
- $10M total Seed · 2026-03 third-party fix
- Freshness
- 2026-10-29 fix
- Layer
- agent-infra repo fix
- Licensing
- open-core (estimate) fix
Memory taxonomy
- storage
- parametric
- retrieval
- parametric-recall
- persistence
- parametric-permanent
- update
- parametric-edit
- unit
- weight
- governance
- opaque
- conflict
- none
When to use
Optimised for: endogenous in-model memory (no external bank)
Anti-fit: searched not found
Pros & cons
Pros
Productivity-focused memory layer with calendar / email integrations.
Cons
Niche to productivity workflows; not general-purpose.
Claims & capabilities
~$10M seed (March 2026, L2F Light Source + Zhongding Capital). Claims exponential improvement in memory-training efficiency vs external-store approaches.
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
- not documented publicly
- A2A
- not documented publicly
- OpenTelemetry
- not documented publicly
Alternatives to MemoraX AI → live options ranked by shared design, licence, hosting and MCP support
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 (24 cited facts, 17 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Endogenous / in-model memory (no external bank) third-party
- What it does & what's distinct Shenzhen. Argues attaching external vector stores is architecturally flawed; trains memo… vendor
- Claims / benefits ~$10M seed (March 2026, L2F Light Source + Zhongding Capital). Claims exponential improv… third-party
- Modalities text third-party
- Created 2026-03 third-party
- Latest release MemoraX Code 0.1.21 (npm @memorax/memorax-code, 2026-10-21) registry
- License MIT (MemoraX Code client); hosted memory platform proprietary repo
- GitHub memorax-ai/memorax-code (MemoraX Code, MIT); org github.com/memorax-ai also hosts Script… repo
- Mindshare 1 press ~20 (Shenzhen s jobs third-party
- Funding $10M total Seed · 2026-03 third-party
- HQ China (Shenzhen, Nanshan district) third-party
- Founders / pedigree Shenzhen-based; founder He Jianye (Tianjin University professor) third-party
- Memory volume / scale no scale claim third-party
- Versioning snapshots third-party
- Schema evolution schema locked third-party
- Contradiction handling LLM resolves and overwrites third-party
- Integration count 0 third-party
- Orchestration agnostic third-party
- Validated verticals China market, agent developers vendor
- Optimised for endogenous in-model memory (no external bank) vendor
- Pros Productivity-focused memory layer with calendar / email integrations. third-party
- Cons Niche to productivity workflows; not general-purpose. third-party
- Project & sources 36kr coverage third-party
- taxonomy-three-layer-rollup agent-infra repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-29
- Homepage “https://eu.36kr.com/en/p/3785834045583875” → “https://www.memoraxai.com/”
- License “not applicable — not OSS” → “MIT (MemoraX Code client); hosted memory platform proprietary” source
- Latest release “not applicable — not OSS” → “MemoraX Code 0.1.21 (npm @memorax/memorax-code, 2026-10-21)” source