# Spatial reasoning

> Spatial reasoning is the process of turning physical context into structured decisions. Rowe combines geometry, object ids, raster views, sensor readings, temporal history, and goals to reason about the state of a scene.

Source: https://fourechelon.com/docs/spatial-reasoning · Machine-readable: https://fourechelon.com/openapi.json · Index: https://fourechelon.com/llms.txt

## What the term means here

Turning physical context into structured decisions. Rowe combines geometry, object identity, raster views, sensor state, temporal history, and goals to answer questions about a scene — whether something fits, whether it is stable, what changed, what happens next.

## Why geometry has to survive the trip

A vision-language model sees a rendered picture of your part. Rowe sees the part. Solids, faces, dimensions, and constraints arrive as structured input, so an answer about a 4 mm radius is computed from a 4 mm radius rather than inferred from pixels.

## Pick the task, then the input

Start from the question you are asking — fit, stability, placement, comparison, prediction, optimization — and send the smallest scene that can answer it. Each model guide lists the fields its task actually depends on.