Exabase is #1 on the leading AI memory benchmark

Read more

Exabase is #1 on the leading AI memory benchmark

Read more

Memory API

Memory API

Everything you need to build AI that actually remembers.

Semantic search, context retrieval, sub-300ms response times.

Everything you need to build AI that actually remembers.

Semantic search, context retrieval, sub-300ms response times.

Read the docs

Get started in <1 minute

Read the docs

Over

Over

20,000,000

20,000,000

files, workspaces, and memories created

files, workspaces, and memories created

Problem

Agents can't remember anything.

Your agent forgets everything between sessions.


Users repeat themselves. Your copilot suggests the same rejected patterns. Support bots ask questions they've already been answered. You try storing conversation logs and running vector search, but it falls apart with conflicting facts, stale context, and noise.


Memory is harder than it looks.

Solution

Memory that actually works

Exabase Memory isn't just storage and retrieval. It tracks relationships, handles contradictions, and evolves over time.


Customer support example:

A user says they are on the free plan. They upgrade to enterprise two weeks later. The support agent still retrieves and reasons from their original plan tier, offering limitations that no longer apply and missing features they now have access to.


Exabase Memory knows the relationship changed.

Datapoint

Exabase M-1 memory system is benchmarked as #1 in the world on the leading benchmark.

*Exabase Research, May 2026

*Exabase Research, May 2026

Automatic memory extraction

import { Exabase } from "@exabase/sdk";
const api = new Exabase({
  apiKey: process.env.EXABASE_API_KEY,
});
const memoryJob = await api.memories.create({
  source: 'text',
  content: 'My favourite ice cream flavor is pistachio.',
});

Send your agent's conversations to Exabase. It extracts facts, preferences, and events automatically.

No need to decide what's worth remembering.

Or set memories directly

const memoryResponse = await api.memories.create({
  source: 'text',
  infer: false,
  content: 'User likes pistachio ice cream.',
});

Write specific memories through the API when you need precise control.

Directly set any text content as a memory.

Dynamic self-optimising memory.
Managed for you.

A self-organizing memory engine, purpose-built from the ground up.


Not a vector database. A living, addressable network that interconnects concepts, learning and evolving with every interaction.


Add memories. Store facts, preferences, events. Anything you learn about a user.


Exabase builds a dynamic ontology: tracking relationships, resolving contradictions, evolving as new information comes in.


Search semantically. Find relevant memories by meaning, not keywords.


Get context. One call returns everything relevant for a task, formatted and ready for your prompt.


We handle extraction, indexing, relevance scoring, and retrieval. You focus on building.

Dynamic self-optimising memory.
Managed for you.

A self-organizing memory engine, purpose-built from the ground up.


Not a vector database. A living, addressable network that interconnects concepts, learning and evolving with every interaction.


Add memories. Store facts, preferences, events. Anything you learn about a user.


Exabase builds a dynamic ontology: tracking relationships, resolving contradictions, evolving as new information comes in.


Search semantically. Find relevant memories by meaning, not keywords.


Get context. One call returns everything relevant for a task, formatted and ready for your prompt.


We handle extraction, indexing, relevance scoring, and retrieval. You focus on building.

Production-ready

Private by design

Security-first

Scalable

Production-ready

Private by design

Security-first

Scalable

Production-ready

Private by design

Security-first

Scalable

Production-ready

Private by design

Security-first

Scalable

Memory API at a glance

Endpoint

Method

Description

/v2/memories

POST

Add a memory

/v2/memories/{id}

GET

Get a memory

/v2/memories/{id}

PUT

Update a memory

/v2/memories/{id}

DELETE

Delete a memory

/v2/memories/search

POST

Memory semantic search

/v2/memories/jobs/{jobId}

GET

Get memory job status

View full API reference →

Memory that improves accuracy

Memory isn't just about personalization.
It's about reliability.

Agents hallucinate when they either lack or have too much context.

They make up facts, forget constraints, ignore preferences the user already stated. Memory grounds your agent in what actually happened.

Exabase Memory updates continuously as new information comes in. Contradictions get resolved. Stale context gets refreshed. Your agent stays accurate, not just personalized.

28% fewer hallucinations when agents have access to reliable memory and context.*

*Based on research into context-augmented AI systems.

Works with everything

Model-agnostic. Framework-agnostic.

SDKs

Python and JavaScript clients. Or call the REST API directly.

CLI

Use Exabase from the command line. Works with Claude Code, shell scripts, or anywhere you can run a terminal.

MCP support

Connect to Claude Desktop, Cursor, Windsurf, Continue, and any MCP-compatible tool.

Ready for scale

Fast

Sub-300ms retrieval. Infrastructure that won't slow your agent down.

Secure

Encrypted in transit (SSL) and at rest (AES-256).

CASA certified.

Reliable

99.9% uptime. Built on Exabase's consumer-grade scaled infrastructure.

Example projects

View all

Build a GitHub PR Knowledge Base with Exabase

Build a Memory-Powered Personal Assistant with Exabase

Build a Sales Memory Copilot with Exabase

Deciding?

Ask your favourite AI about Exabase:

Deciding?

Ask your favourite AI about Exabase:

Ship your first app in minutes.

Start for free

Read the docs

Ship your first app in minutes.

Join the waitlist

Read the docs