> ## Documentation Index
> Fetch the complete documentation index at: https://docs.autosage.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Set up durable memory

> Enable durable memory and tune how the assistant captures and recalls facts across conversations.

Durable memory lets an assistant carry facts, preferences, and decisions from one conversation into future, separate ones. It's optional and off by default. You enable and tune it with two tenant settings, then optionally scope it further per knowledge base.

## The two tenant settings

<Steps>
  <Step title="Choose a mode">
    The mode decides whether durable memory is active at all.

    | Mode         | Behavior                                                                   |
    | ------------ | -------------------------------------------------------------------------- |
    | **Disabled** | No durable memory. The assistant only remembers within each conversation.  |
    | **Durable**  | Durable memory is on. The assistant can recall facts across conversations. |
  </Step>

  <Step title="Choose a retention level">
    When the mode is **Durable**, retention decides *how* new facts get into memory.

    | Retention         | Behavior                                                                                                          |
    | ----------------- | ----------------------------------------------------------------------------------------------------------------- |
    | **None**          | Recall only. The assistant uses what's already stored but adds nothing new.                                       |
    | **Explicit only** | Stores a fact only when a user explicitly asks it to remember something. See [Memory commands](/memory/commands). |
    | **Automatic**     | Also captures relevant facts from conversations on its own, in addition to explicit requests.                     |
  </Step>
</Steps>

## Mode × retention at a glance

The two settings combine to give you precise control over reading and writing memory:

| Mode     | Retention     | Recalls existing memories? | Stores new memories?              |
| -------- | ------------- | -------------------------- | --------------------------------- |
| Disabled | —             | No                         | No                                |
| Durable  | None          | Yes                        | No                                |
| Durable  | Explicit only | Yes                        | Only when explicitly asked        |
| Durable  | Automatic     | Yes                        | Yes, automatically and on request |

## Scoping further with a knowledge base

Each knowledge base has its own [memory mode](/knowledge-bases/memory-modes) — **Disabled**, **Subject**, or **KB-restricted**. A knowledge base can only *further restrict* the tenant setting, never widen it.

<Info>
  If the tenant has durable memory turned off, a knowledge base cannot turn it on. The tenant setting is always the ceiling, and the knowledge base setting narrows from there.
</Info>

## What happens when memory is enabled

Once durable memory is active, recall and storage happen quietly in the background:

* On each new message, relevant memories are automatically retrieved and added to the assistant's context.
* The assistant indicates when a memory was used, so it's clear where recalled context came from.
* New facts are stored according to your retention level.

<Note>
  Recall and storage are best-effort. A memory-service failure does not block a normal answer; if nothing relevant is found, the assistant simply answers without recalled context.
</Note>

## Choosing your settings

<Tip>
  Start with **Explicit only** if you want users to stay in control of exactly what gets remembered. Choose **Automatic** for assistants that should learn preferences and context over time with no extra effort from the user.
</Tip>

## Next steps

<CardGroup cols={2}>
  <Card title="Memory commands" icon="wand-magic-sparkles" href="/memory/commands">
    Store facts conversationally with natural phrasing.
  </Card>

  <Card title="Manage memories" icon="list-check" href="/memory/managing">
    View, search, and delete what's been remembered.
  </Card>

  <Card title="Memory modes" icon="book" href="/knowledge-bases/memory-modes">
    Restrict memory for a specific knowledge base.
  </Card>

  <Card title="Memory overview" icon="brain" href="/memory/overview">
    Revisit how the memory levels fit together.
  </Card>
</CardGroup>