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

# Custom models

> An embedding model fine-tuned on your data, acceptance-gated against a frozen evaluation set of real queries, and promoted only on a measured win.

Conare builds your company's own retrieval stack: an embedding model fine-tuned on your data, evaluated against your real queries, served with its own reranker and the database where your data lives.

This is not an API you self-serve into — it's an engagement, because the work is specific to your data. Here's exactly how it runs.

## How the engagement works

<Steps>
  <Step title="Connect your data">
    Your corpus flows in through the [connectors](/integrations/overview) or the API — the same ingestion paths as the rest of the platform. Your data trains **your** model; it is not pooled into anyone else's.
  </Step>

  <Step title="Freeze the evaluation set">
    We build a golden set from your real queries and judged outcomes — frozen before training starts, so the target can't drift to flatter the model. This set is the acceptance gate for everything that follows.
  </Step>

  <Step title="Train and evaluate">
    The model is fine-tuned on your distribution and scored against the frozen set on the metrics that decide retrieval quality: recall on your queries, ranking position of the right answer, and calibrated no-answer behavior.
  </Step>

  <Step title="Promote only on a proven win">
    A candidate ships only when it beats the incumbent on your evaluation set. No vibes-based upgrades: if it doesn't win on your queries, it doesn't serve.
  </Step>

  <Step title="Serve pinned">
    The winning model serves your traffic with an explicit version pin — the same [never-silently-substituted guarantee](/embeddings/api#model-pinning) as the hosted API — alongside a reranker tuned to the same distribution. Ongoing maintenance keeps the model current as your data evolves.
  </Step>
</Steps>

## What you end up with

* An embedding model that is measurably better **on your queries** than the generic API you use today — with the evaluation receipts to show it.
* A matched reranker, so first-stage recall and final ordering improve together.
* Serving in the same platform as your data — no cross-vendor pipeline to operate.

## Start a conversation

Custom models are scoped per engagement — corpus size, query volume, and evaluation criteria shape the work.

<Card title="Book a call" icon="calendar" href="https://cal.com/amurzin/30min">
  30 minutes with the founder: bring a description of your corpus and a handful of real queries where retrieval disappoints you today.
</Card>