Redis expands Iris with six capabilities for AI agents

Redis has expanded Redis Iris, its real-time context engine for AI agents, with six capabilities that supply the memory, context, and messaging agents need to operate in production.

According to the release, AI models and infrastructure are now good enough for enterprise use, yet few teams report repeatable, production-level results with agents. Agents often lack the right context at the right moment, and those that do deliver value tend to do so with over-stuffed prompts, redundant LLM calls, and brittle custom integrations that cost more in tokens and time than they return.

Redis Iris now includes six capabilities that can be used together or independently. Intent Router, which is generally available, matches the meaning of a request rather than its wording and sends it to a remembered FAQ, an agent skill, a human, or an LLM. It also supports agent and model routing, guardrails, and “fuzzy” if-statements, so decisions are not limited to true or false.

Agent Playbooks, also generally available, let teams separate answers that are stable and reusable from those requiring current data and a fresh run of a known procedure. Playbooks become more thorough and accurate with repeated runs, which reduces resolution times and token spend. Agent Memory, generally available as well, builds memories from user actions, documents, and other semi-structured data, then organizes them in hierarchies with inspection, audit, and provenance.

Context Gateway is generally available and gives agents a governed entry point to connected context sources. Lens and Agent Mesh are in beta; Lens traces each outcome to the skill, route, or FAQ that served it, and Agent Mesh buffers and reconnects agent streams when a connection drops. Because Iris runs on Redis, teams can reduce the custom integration work that typically connects retrieval, memory, routing, and messaging.

Customers have used Redis Iris in production since its May 2026 launch. “Everything you see in Paradigm is stored in Redis. It is the memory and orchestration layer that lets our agents work in real time, at a scale that wasn't possible before,” said June Lee, CTO, Paradigm. A live first look with the engineering team is planned for November 12 (Americas) and November 17 (EMEA).