Weaviate Makes Engram Generally Available in Weaviate Cloud

Weaviate said it made Engram generally available, a managed memory and context service for agentic applications. The release positions Engram as a way to address reliability and continuity limits for agent systems when work spans multiple sessions or agents.

As organizations moved AI agents into production, the company cited a recurring constraint: agents could not reliably remember users across sessions, learn from feedback, or share context when a task ran across several agents. It also said teams relied on workarounds such as replaying full conversation histories, hand-curating files, or building custom memory systems.

Engram treated memory as first-class infrastructure alongside storage and retrieval. It kept memory actively maintained rather than accumulating ever-growing context by running asynchronous pipelines to extract facts, reconcile them with existing knowledge, and handle deduplication and updates. The company said applications handed off raw events while memory was built in the background, and memories were served back through Weaviate’s hybrid search.

Scoping isolated memories per project, per user, and per property so context reached only the right caller. The company also referenced ready-made templates for common use cases including personalization, continual learning, and multi-agent shared state, with the option to customize projects. “Memory is the difference between an agent that answers a question and an agent that gets better at its job,” said Bob van Luijt, CEO and co-founder, Weaviate. “With Engram, every team gets a production-grade memory layer without building or operating one themselves on infrastructure they already trust.” Engram was available in Weaviate Cloud, including a free tier with 1,000 pipeline runs per month, with paid plans starting at $45 per month.

Provided by Globe Newswire on behalf of Weaviate. Click to read original content.