What is Prompt?
The written instruction given to an AI model. In architectural rendering, strong prompts name the building type, materials, lighting and camera — e.g. "single-storey concrete house, dark timber cladding, golden hour, eye-level view".
Updated 4 November 2026
A prompt is the text instruction that tells an AI model what to produce or how to change an image. In architectural rendering it works like a short brief to a visualiser: what the building is, what it is made of, what light it sits in, and from where the camera looks.
Why it matters
The model can only act on what you say or show. A clear prompt gives predictable results and saves credits and time; a vague one gives generic, flat images. When you also provide an input image, the prompt controls the finish rather than the structure, so short and specific descriptions work best.
Example
Weak: "nice modern house". Stronger: "single-storey concrete house, dark timber cladding, golden hour, eye-level view, light mist". The second names the type, materials, lighting and camera, so the model has far less to guess.
Common mistake
Stuffing a prompt with contradictory or repeated adjectives, or describing geometry that the input image already defines. Keep it to the few things that matter, and change one thing per iteration so you can see what each change does.
Practical tips
Keep a personal library of prompts that worked for particular building types and light conditions, and reuse them as templates. Describe what you want to see, not what you want to avoid, as negative wording often backfires. Use plain nouns for materials and a time of day for light. Test one change at a time and keep notes so improvements are repeatable.
In Archome AI
Archome AI has a guide with ready-to-use wording: see 40 render prompts, and apply them in Imagine AI or any image-based tool.