MariaDB Enterprise Platform 2027 unifies agent memory and AI access in one data layer
MariaDB Enterprise Platform 2027 combines Agent Memory, Semantic Cache, AI Retrieval and coding-agent plugins with transactions, analytics and vector search on one platform.
AT A GLANCE
Key points
- Agent Memory preserves state across sessions.
- Semantic Cache aims to reduce repeated model calls.
- AI Retrieval creates read-only SQL from documents and live data.
What is changing?
MariaDB made Enterprise Platform 2027 generally available on November 4. The release adds persistent memory, semantic caching, data access and development tools for AI agents alongside transactions and real-time analytics. It can run on premises, in a cloud or through MariaDB Cloud.
Agent Memory lets long-running agents preserve state across sessions. Semantic Cache aims to reduce latency and token costs by reusing stored answers for similar questions. AI Plugins provides MariaDB expertise to coding tools such as Claude Code and Codex, while AI Retrieval produces read-only SQL over documents and live operational data.
The platform includes MariaDB Enterprise Server 12.3, Exa, GridGain, MaxScale and management components. AI Retrieval is scheduled for public preview on MariaDB Cloud later this month; production deployments still need to audit access permissions and generated SQL.
Explore the original sources
Source date: 4 November 2026.
Published by AI Radar: 5 Nov 2026, 00:04 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…