R

Redis Agent Memory

API service

by Redis

A memory service for AI agents that keeps each conversation as ordered session events with automatic summarisation, extracts durable facts into long-term memory in the background, and retrieves them across sessions by semantic, keyword or hybrid search.

Find alternatives to Redis Agent Memory

What it does

• Agent orchestration
• Vector search

Features 6

• Working memorySession-scoped working memory storing each conversation's live messages.
• Cross-session extraction policiesConfigurable LLM-based policies extract facts, preferences and episodic events and store them as vectors.
• Automatic memory managementSummarizes and trims session history and promotes high-signal facts to long-term memory.
• Contextual groundingResolves pronouns and references in conversations.
• DeduplicationPrevents duplicate memories with content hashing.
• REST API, MCP server and Python clientMultiple interfaces for accessing agent memory.

Sources

Pricing

Not found · 10 Oct 2026

The page shows "Start for free" and "Talk to sales" with no rate; redis.io/pricing/ does not name the service. "~$0.07 per session" in the banner is a benchmark result.

Description

Official documentation · 10 Oct 2026

"Redis Agent Memory is a memory service for AI applications. It stores ordered conversation events in session memory and durable information in long-term memory." Long-term memory is searchable by "semantic, keyword, or hybrid search".

What it does

Official documentation · 10 Oct 2026

Sold within

Not found · 10 Oct 2026

One of the five tools "Iris is composed of"; created as a managed service from the Redis Cloud console. No page read says how it is billed. Read: the Agent Memory page and docs, the Iris docs and redis.io/pricing/.

Name

Official documentation · 10 Oct 2026

"A managed memory layer that gives agents intelligent short-term memory and persistent context across conversations."

URL

Official documentation · 10 Oct 2026

Deployment

Official documentation · 10 Oct 2026

Redis Cloud managed service, or self-managed on Kubernetes with Redis Software (private preview).

Also from Redis

• Redis Search API service

  Vector search and hybrid (vector + exact-match) query engine built into Redis, marketed as "the world's fastest vector database" for GenAI apps and RAG pipelines at billions-of-vectors scale.

• Redis Iris Infrastructure service

  Redis's context engine for AI agents, bundling Context Retriever, Agent Memory, LangCache, Data Integration and Redis Search into one runtime that keeps agents supplied with live business data, memory and cached answers, managed on Redis Cloud or self-managed on Kubernetes.

• Redis LangCache API service

  A semantic cache for LLM calls: before an app calls its model, LangCache embeds the prompt, looks for a semantically similar prompt already answered and returns the stored response, with configurable similarity thresholds, TTLs and eviction, through a REST API.

• Redis Context Retriever API service

  A schema-first context layer that turns a modelled set of business entities, fields and relationships into generated retrieval tools agents call over MCP, so agents read live operational data through governed, row-filtered lookups rather than writing SQL.

• Redis Copilot Developer tool

  An AI assistant in Redis Insight and the Redis docs that answers questions about Redis commands and concepts and, connected to a user's own database, turns plain-language requests into Redis search queries over that data.

Open Redis in the directory

Something wrong here? Send a correction — quote this product id: redis-agent-memory.