Build production-ready, resilient agents. Six new capabilities across retrieval, memory, routing, and messaging.
Redis Iris capability
Managed memory to store state & context that persists.
Give agents intelligent short-term memory and persistent context across conversations in a single, fully customizable managed memory layer.
Agent Memory
Automatically summarize and trim session history, promote high-signal facts from short-term to long-term memory, and use semantic & metadata retrieval so agents always see the most relevant context without custom promotion or pruning logic.
Automatic extraction
Pull important facts from every agent conversation.
One memory, multiple channels
Share one memory across channels, from chat to email to voice, so context carries over and every interaction picks up where the last one left off.
Deduplication
Prevent duplicate memories with content hashing.
Multiple interfaces
Work with REST API, MCP server, Python client.
Authentication
OAuth2/JWT, token-based, or disabled for development.
Multi-agent
Memories are shared, consistent, and scoped between agents, so each can build on what others have learned without duplicating work.