Couchbase updates AI platform with Amazon Bedrock integration

Couchbase expanded its AI Data Plane with Amazon Bedrock integration with semantic caching, managed agent memory and a managed model context protocol server on Amazon Web Services. The company also added vector search to Couchbase Operational Insights.

The updates along with its moves to create a single governed platform are part of Couchbase's plan to become an operational data platform for AI and compete with the likes of MongoDB.

• Couchbase plots course to be broader AI data platform
• Constellation ShortList™ Hybrid-Cloud and MultiCloud NoSQL Databases

Couchbase said its platform can lower token consumption by up to 45% and simplify AI deployments. According to Couchbase, the additions to its platform extend its foundation with managed cloud services and lower token costs.

Holger Mueller, an analyst at Constellation Research, said:

"When platform vendors move close to AI platforms, the customers win. Today's it's Couchbase that integrates better with AWS Bedrock - and the customers win - better performance and significantly lower token costs lower the ROI hurdle for AI automation and thus power Enterprise Acceleration."

"Customers building production AI agents at scale need a converged data layer, but most are still running three or four separate systems for memory, vectors, operational data, and analytics," said Barry Morris, chief product and strategy officer at Couchbase. "What we’re shipping now collapses that stack."

According to Couchbase, the company's move to unify key-value, SQL, full-text search, vector search and real-time columnar analytics on a single data set gives AI agents operational and historical context in one call instead of hitting separate systems.

Key items in the Couchbase update include:

• Amazon Bedrock integration that will expand its model selection and provide other tools including vectorization over private networking and serverless management.
• Couchbase Agent Memory will be available as a managed cloud service on AWS. Agent Memory was launched in June.
• Couchbase Operational Insights now supports vector search alongside structured analytical queries.