MemoryLake
To transition AI infrastructure from 'data-centric' to 'memory-centric' by providing a persistent multimodal memory layer for AI agents.
At a Glance
- AI Agent Developers
- Enterprise Data & Analytics Teams
- Individual AI Power Users
AI Tools by MemoryLake
(1)
MemoryLake
Cross Model Memory Service
Discussions
No discussions yet
Be the first to start a discussion about MemoryLake
Latest News
MemoryLake Dominates AI Memory Benchmark with 94.03% Accuracy
MemoryLake Launches Lobster Edition for OpenClaw Agents
Zhibian Technology Releases Its First Multimodal Memory Lake: MemoryLake
MemoryLake (Zhibian Tech) Raises Tens of Millions in Angel Round led by Hillhouse and Lightspeed
Products & Services
A multimodal memory passport that travels across AI platforms (ChatGPT, Claude, OpenClaw, etc.), providing persistent long-term memory.
Proprietary multimodal VLM (Vision-Language Model) engine for parsing complex documents, Excel layouts, and visuals for memory indexing.
Human-like decision-making AI agent capable of complex reasoning and autonomous operations.
Market Position
Positioned as the 'World's First Hyperscale AI Memory Platform,' prioritizing architectural privacy and multimodal accuracy over simple session logging.
Leadership
Founders
Zhan Chaoqun (占超群)
Known as Li Zhe (离哲). Former Alibaba Cloud Researcher and Head of Database OLAP Product Department. Chief architect for Alimama's Damo Panel and Hangzhou 'City Brain' project. Headed the development of AnalyticDB (ADB).
Wei Chuangxian (魏闯先)
Co-founder of Zhibian Tech (MemoryLake). Expert in database and AI infrastructure.
Executive Team
Zhan Chaoqun (Li Zhe)
Founder & CEO
Former P10 executive at Alibaba Cloud; led the core data infrastructure for Alibaba's cloud-native data warehouse.
Julian Zhou
Head of Products
Product leader in Data and AI; former product manager at Alibaba Cloud Database team. Holds over 50 patents in data systems.
Board of Directors
Founding Story
Founded by Zhan Chaoqun (Li Zhe), former head of Alibaba Cloud's OLAP products, who observed that LLMs lacked session-persistent memory. He aimed to build a 'second brain' for AI to enable agents with human-like recall and cognitive accumulation.
Business Model
Revenue Model
Freemium subscription model supplemented by on-demand usage credits ($3.125 per 1M tokens).
Pricing Tiers
300,000 tokens per month. Best for trying the product.
6,200,000 tokens per month. Best for individual or small team usage.
66,000,000 tokens per month. Best for heavy usage and team-scale workloads.
Target Markets
- AI Agent Developers
- Enterprise Data & Analytics Teams
- Individual AI Power Users
- Cross-session AI memory persistence
- Enterprise-grade agent memory
- Long-context data indexing and retrieval
- Multimodal document understanding
- OpenClaw ecosystem
- Early enterprise adopters in tech and finance sectors