---
title: Adaptation and fine-tuning
description: How a copy of Avra RFM learns from your database or your task, when that pays, and why the copy stays exclusive to your workspace.
---

In-context prediction is the default: the shared model reads your history when a prediction is made, and no weights change. Where it pays, a copy of the RFM learns from your data instead, in one of two ways.

| Mode | What happens | Effect on the model |
|------|--------------|---------------------|
| **In-context prediction** | Your history is read as context when a prediction is made | No weights change |
| **Adaptation** | Self-supervised learning on your database | A copy exclusive to your workspace |
| **Fine-tuning** | Learning one particular task | A copy exclusive to your workspace |

<svg class="fig" viewBox="0 0 640 250" role="img" aria-label="On the left, the shared Avra RFM. On the right, inside your workspace, your history, which the shared model reads as context, and two copies of the model, one adapted to your database and one fine-tuned on a task. Arrows go one way, from the shared model to the copies.">
  <defs>
    <marker id="fig-copies-arrow" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="7" markerHeight="7" orient="auto-start-reverse">
      <path d="M0,0 L10,5 L0,10 z" class="arrow" />
    </marker>
  </defs>
  <rect class="rule dash" x="262" y="18" width="362" height="222" rx="12" />
  <text class="small title label" x="278" y="40">Your workspace</text>
  <rect class="solid" x="24" y="102" width="170" height="56" rx="10" />
  <text class="mid on-solid title" x="109" y="126">Avra RFM</text>
  <text class="small mid on-solid" x="109" y="144">shared, pre-trained</text>
  <rect class="card" x="300" y="54" width="300" height="38" rx="6" />
  <text class="small" x="316" y="77">Your history</text>
  <text class="small label end" x="586" y="77">read as context</text>
  <path class="edge" d="M300,73 C250,73 236,104 198,116" marker-end="url(#fig-copies-arrow)" />
  <rect class="soft" x="380" y="116" width="220" height="44" rx="8" />
  <text class="small title" x="396" y="135">Adapted copy</text>
  <text class="small label" x="396" y="151">learned your database</text>
  <rect class="soft" x="380" y="176" width="220" height="44" rx="8" />
  <text class="small title" x="396" y="195">Fine-tuned copy</text>
  <text class="small label" x="396" y="211">learned one task</text>
  <path class="edge-accent" d="M194,136 C280,136 300,138 376,138" marker-end="url(#fig-copies-arrow)" />
  <path class="edge-accent" d="M194,146 C280,160 300,198 376,198" marker-end="url(#fig-copies-arrow)" />
  <text class="small label mid" x="300" y="128">copy</text>
</svg>

The shared model reads your history as context without changing. An adapted or fine-tuned copy is made inside your workspace, and it's never merged back into the shared model.

## Adaptation

The copy learns from your database itself, without labels: how your tables connect and what their values look like. Every question you then ask over that database starts from a model that already knows it, so adaptation helps most when you'll ask many questions of the same database.

## Fine-tuning

The copy learns one task from your labels: the outcome you care about most, at the horizon you need. Fine-tuning helps most when one prediction carries a decision that matters enough to justify its own model. It also opens up other kinds of targets, such as one category among several. See [What you can predict](/pt-BR/rfm/what-you-can-predict#kinds-of-targets).

## When to use them

Start in context. It needs no training run, so you learn whether a question is worth pursuing before you train anything. Adapt or fine-tune when an evaluation on your history shows the copy improves on in-context results by enough to justify maintaining it, because a copy is a model of its own to evaluate, release, and keep current.

An adapted or fine-tuned copy is evaluated and released like any other version. See [Evaluation](/pt-BR/rfm/evaluation) and [Model versioning](/pt-BR/core/model-versioning).

## Your copy

An adapted or fine-tuned copy is exclusive to your workspace and never serves another customer. It's a one-way fork: it's never merged back into the shared model, so nothing about your business reaches anyone else's model.

The copy is derived from Avra's model and licensed for use in your workspace, not delivered: its weights are never handed over as usable files. It's stored where you run. In Avra's cloud, that's inside your isolated workspace. In your AWS account, the training job that produced it writes it, encrypted, to your S3.

## How to run them

In Avra's cloud, Avra runs adaptation and fine-tuning in your workspace through [Avra Core](/pt-BR/core). In your AWS account, they're training jobs on Amazon SageMaker, and inference jobs then use the copy they produced. See [Avra RFM on AWS SageMaker](/pt-BR/deployment/aws-sagemaker).

**[Evaluation](/pt-BR/rfm/evaluation)**

How a copy proves it's worth maintaining

**[Data privacy & compliance](/pt-BR/security/data-privacy-and-compliance)**

What stays yours, and what Avra learns from operating the platform
