Essential for context engineering

Akka Memory provides durable, in-memory, and sharded data for AI agents, enabling superior context engineering—structuring and optimizing agent information to enable effective and reliable performance of complex tasks across multiple interactions over time. Akka Memory provides your AI with comprehensive short-term and long-term recall, essential for coherent, personalized, and informed interactions.

Prompts Manage system and user prompts outside of code with versioning and rollback.
User input Integrate queries and requests with semantic knowledge.
Short-term memory Maintain situational awareness and conversation history through past interactions between users, agents, and tools.
Long-term memory Persist semantic knowledge, skills, and retrieved data across users, sessions, agents, and systems.
Knowledge retrieval Access dynamic information through connectors to vector databases, transactional databases, brokers, or APIs.
Tool schemas Access dynamic information from external systems through functions, tool calls, and MCP servers.
Workflows Guide agents through complex processes with workflows that persist their call stack on every step.
Compaction Optimize information for LLM context windows and token consumption.
Evaluation events Capture every agent event and context sent to the LLM to evaluate context effectiveness

Durable event-sourced memory

The state is tracked sequentially as a series of events and persisted transparently in an event journal without requiring developer intervention. Snapshots and event replay enable resilience of memory from any network or hardware failure.  
empty-recovered-state-agent
subscribe-memory-alt

Subscribe to memory

Agents can subscribe to any event or changes in memory using Akka Streaming, enabling adaptive behavior in response to real-time environmental changes. 

Akka Memory tracks state with an event-driven architecture, recording every change as an immutable fact. These events can be replayed for recovery, replicated for resilience, broadcast for real-time updates, or projected into custom read models. 

Developers can create multi-memory queries using Views that optimize read access or define custom subscriptions to memory change events through Akka Streaming.

Sharded and in-memory for <10ms writes

Data sharding

App data is partitioned across in-memory, durable nodes.
Akka routes user requests to the correct instance.

sharded-data

Data rebalancing

Data shards are re-balanced as the
number of runtime nodes changes.

data-rebalancing-1

Traceable and auditable

Capture, review, and replay every memory interaction. For agents, trace user, tool, and LLM conversations. For workflows, trace transitions, and call stacks for long-running processes. Akka stores memory events in an encrypted, tamper-evident record for audit, compliance, and deep system analysis.

memory-event-journal-alt
transparent-memory-alt

Intrinsic to agents

Agents, workflows, and streams get memory automatically, without requiring developers to program state or persistence logic. Developers can also create their own memory objects that are shared across users, sessions, agents, and systems. 

Part of the platform

Akka Memory is part of the Akka Agentic Platform, which provides everything needed to build, run, and evaluate agentic systems. This includes structured, goal-directed agents with Akka Agents, durable execution with Akka Orchestration, and real-time data handling with Akka Streaming.

Akka Memory is high-performance, durable, and accessible both locally and across regions.

akka-platform

Additional resources

Akka Memory: Durable, in-memory, and sharded data

Akka Memory provides durable, in-memory, and sharded data for AI agents.
Learn More

Demo: Temperature monitoring agent

In this video, Akka engineer Andrejz showcases a temperature-monitoring solution designed to gather, consolidate, and analyze IoT sensor data.
Learn More

Building a real-time video AI service with Google Gemini

Learn how we built an agentic AI service for real-time video with Google Gemini, overcame API challenges with Akka, and scaled to 1000s of TPS in Google VPC.
Learn More

When AI Needs an SLA

Get Started