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

# Models

> Pre-trained models specialized per job

AiQL runs a family of pre-trained models, each specialized for one job. Convert uses layout and OCR. Shape and extract use pipeline language models. Queries use answer models over embeddings of the knowledge layer. This handbook does not name the checkpoints.

<Columns cols={2}>
  <Card title="Layout" icon="table-cells" href="/products/models-layout">
    Finds regions on each page before any text is read.
  </Card>

  <Card title="OCR" icon="file-image" href="/products/models-ocr">
    Reads those regions, with dedicated passes for tables, formulas, charts, and seals.
  </Card>

  <Card title="Pipeline" icon="sitemap" href="/products/models-pipeline">
    Infers or applies a schema, then extracts entities and relationships.
  </Card>

  <Card title="Answers" icon="comments" href="/products/models-answers">
    Reasons over the workspace knowledge layer in ask, reason, or build mode.
  </Card>

  <Card title="Embeddings" icon="circle-nodes" href="/products/models-embeddings">
    Vectors for queries, schema names, and instances. Retrieval fuses exact, lexical, and semantic match.
  </Card>
</Columns>