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

# How Slowave differs from RAG and other options

> Slowave is a living memory, not a search index with a model on top. See how it differs from RAG, rules files and hosted memory services.

## Slowave is not a search index with a model on top

Many memory tools work like a library. They store text, fetch the closest matches for a question, and hand them to a language model. This approach, often called RAG, is excellent for searching documents and knowledge bases.

Slowave does a different job. It is a **living memory** that changes as your agent works.

![A typical storage and retrieval setup stores text, fetches the closest matches and hands them to a language model, with no learning. Slowave is a loop: the agent writes memories, receives relevant ones, reports what helped, and background upkeep strengthens, fades or replaces memories without an LLM.](https://mintcdn.com/slowave-ai/hrOsN_VNHRWDj6ew/images/diagrams/not-rag-mobile.svg?fit=max\&auto=format\&n=hrOsN_VNHRWDj6ew\&q=85\&s=763fc865765f2f73688fab38e388c566)

![A typical storage and retrieval setup stores text, fetches the closest matches and hands them to a language model, with no learning. Slowave is a loop: the agent writes memories, receives relevant ones, reports what helped, and background upkeep strengthens, fades or replaces memories without an LLM.](https://mintcdn.com/slowave-ai/hrOsN_VNHRWDj6ew/images/diagrams/not-rag.svg?fit=max\&auto=format\&n=hrOsN_VNHRWDj6ew\&q=85\&s=b46710f61e29aeb73d1184f719cb3b26)

A one-way store-and-fetch pipeline versus a memory that learns from use.

> [card] It learns from use
>
> Your agent reports which memories helped. Helpful ones come back more often. Unhelpful ones fade.

> [card] It keeps itself current
>
> A newer decision replaces an older one. A plain store would return both and leave your agent to guess.

> [card] [It captures how, not only what](https://slowave.ai/procedures.md)
>
> Besides facts, it saves methods that worked, as [procedures](https://slowave.ai/procedures.md).

> [card] Its upkeep needs no LLM
>
> Tidying, ranking and recall run locally with no LLM calls. A small local search model handles matching.

Your agent's own model still does the reasoning. Slowave handles memory. Your agent handles reasoning.

> [!NOTE]
>
> Slowave and document search work well together. Use search for large reference material. Use Slowave for the decisions, rules and lessons that build up as you work.

## Compared to other ways of giving an agent context

| | Rules files (like `CLAUDE.md`) | Search over stored text (RAG) | Hosted memory services (typical) | Slowave |
| - | - | - | - | - |
| **Who writes memory** | You | You index your content | The service | Your agent. You can inspect it in the dashboard. |
| **Where it lives** | Your repo | Wherever you host it | A cloud service | Your computer |
| **Shared between tools** | Per tool | Depends on setup | Usually | Yes |
| **Gets better with use** | No | Typically no | Varies | Yes, from agent feedback |
| **Replaces outdated information** | By hand | By re-indexing | Varies | Yes |
| **Saves reusable methods** | No | No | Varies | Yes, as procedures |
| **Needs an account or key** | No | Varies | Varies | No account. Your agent keeps its own LLM account. |
| **Best for** | Stable team conventions | Searching documents | Apps with many end users | Your own coding workflow |

> [!TIP]
>
> They work well together. Keep stable team rules in a rules file. Let Slowave handle the decisions and lessons that build up as you work.

## When Slowave is a good fit

* You use one or more coding agents every day.
* You are tired of repeating yourself.
* You want your memory to stay on your own machine.

## When it is not

* You need memory for end users of a product you ship. Slowave is for developers using coding agents.
* You need encrypted storage out of the box.

## Preliminary results

We test whether Slowave returns the evidence needed to answer questions about long conversation histories. An external LLM judge scores the retrieved text.

| Test | Result |
| - | - |
| LoCoMo, 1,534 answerable questions | 71.84% |
| LoCoMo, questions spanning several sessions | 87.04% |
| LongMemEval oracle, knowledge updates | 83.33% |
| LongMemEval oracle, all 500 questions | 65.20% |

Scores are lower for time-based questions (about 55% on LoCoMo, 46% on LongMemEval) and for counting across sessions (53%).

> [!WARNING]
>
> These results are preliminary. They measure whether the right evidence was found, not whether a coding agent finished its task better. LongMemEval oracle has no distractor sessions, so it isn't comparable to LongMemEval-S leaderboards. We do not rank Slowave against other systems.

> [card] [Read the code](https://github.com/slowave-ai/slowave)
>
> Slowave is open source under AGPL-3.0-or-later. A commercial license is available on request.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.