Core concepts
AI rendering
Using an AI image model to turn a sketch, clay render, exported 3D model view, or existing visual into a more developed architectural image, guided by a prompt, references and selection tools. See architectural visualization AI for where it fits in a design workflow.
Architectural visualization (ArchViz)
The broader practice of communicating a design through images, from hand sketches to production 3D renders. AI rendering is one route into ArchViz, alongside traditional and real-time rendering; see the architectural visualization AI guide.
Photorealistic rendering
An image built to read as a real photograph: consistent lighting, material response and depth of field. Moving from a neutral clay render to a photorealistic result is one of the most common AI-rendering workflows.
Diffusion model
The class of AI model behind most current image generation and editing, including the engines available in Rendero. Different diffusion models trade off editing control, reference support and output fidelity differently; see how Rendero chose between them.
Real-time renderer
Software such as Enscape, D5 Render or Twinmotion that renders a 3D model directly, built for walkthroughs and physically accurate, production-grade output. AI rendering tools work from a single exported image instead and are not a replacement; see AI vs traditional 3D rendering.
Inputs and source material
Base image
The single image an AI rendering workflow starts from: a sketch, clay render, exported model view or existing visual. Its quality and content set a hard ceiling on the result; see choosing the right base image.
Clay render
A 3D model rendered with a single neutral material and no textures, so massing, openings and camera composition read clearly before any material or lighting decision is made. See the clay render to photorealistic workflow.
Architectural sketch
A hand-drawn, digital, rough or annotated drawing used as an AI-rendering base image. A sketch with clear proportions, openings and structural elements gives the model more to preserve; see sketch to render AI.
3D model view
An image exported or captured from modeling software (for example a Shaded view from Revit, a ViewCaptureToFile capture from Rhino, or a 2D export from SketchUp) and used as an AI-rendering base image. See the Revit, Rhino and SketchUp guides.
Reference image
A separate image supplied alongside the base image to communicate a visual direction, such as material quality, lighting mood, season or camera style, without needing to describe it precisely in words. See reference images as a design language.
Segmentation
Dividing an image into regions, such as facade, roof, glazing or landscape, so an edit can be targeted at one region without redoing the whole image. Segmentation guides a local change; it does not guarantee unchanged geometry elsewhere. See why segmentation matters for architectural accuracy.
Editing and control
Prompt
The written instruction that tells an AI rendering engine what to change and how, for example a material, lighting condition or atmosphere. A prompt works together with the base image and any references; it does not override what the base image already shows.
Selected-area edit (inpainting)
Regenerating only the pixels inside a chosen region, such as a facade material or a piece of landscaping, while leaving the rest of the image untouched. Rendero uses segmentation to define the selected area before an edit runs.
Region map (selection mask)
A color-coded map that marks which parts of an image belong to which region, used to define a selected-area edit. A region map helps define a selection; it does not by itself prove the resulting edges are unchanged, so the result still needs review. See the published selection example.
Style reference
A reference image used specifically to guide the visual style of a result, such as material palette, light quality or a particular photographic look, rather than layout or geometry. See reference images as a design language.
Upscaling
Increasing an already-generated image to a higher resolution as a separate step after generation. Upscaling uses its own credit charge and does not change the composition or materials already decided in the base result; see the FAQ.
Geometry preservation
How closely an AI rendering result keeps the openings, proportions and structural layout of the original design as the model, sketch or clay render defined them. It is a central quality question for architectural AI rendering, not a guaranteed property of any model; see why control, not just image quality, is what matters next.
Workflow and delivery
Design intent
The design decisions that must stay fixed through a visualization, such as an opening's position, a material family, or a spatial proportion, regardless of how a render is generated. Reviewing a generated image against design intent, not just against how convincing it looks, is the core discipline in AI rendering for architects.
Client-ready visual
An image checked and approved for an actual client or investor presentation, as distinct from an internal concept or mood study. Rendero can support drafts and comparisons at this stage, but every image still needs review against the design before it is shown; see concept to client presentation.
Material study
A round of AI-rendering exploration focused on comparing material or finish options on the same base image, rather than finalizing a presentation image. See what to render fast, and what to decide slowly for which decisions suit a quick study versus a slower review.
Interior rendering
AI rendering applied to interior spaces: layout, furnishing, materials, and light. Interior work follows the same base-image-plus-prompt workflow as exterior rendering, with its own sequencing; see the 6-step AI interior rendering workflow.
Off-plan visualization
Rendering used to market a property before construction is complete, common in investment presentations and pre-sale marketing for new developments. Rendero publishes regional guides for markets where this use case is common, including Vietnam and Indonesia.
Plans and credits
Rendering credit
Rendero's unit of usage: generating or upscaling an image spends credits, at a rate that varies by AI engine, resolution and, for some engines, processing time. A credit balance does not represent a fixed number of finished images once retries and edits are counted; see the pricing FAQ.
Free plan
A Rendero account tier that includes 50 starting credits, granted once at signup, with no monthly refill. It is meant for testing the workflow on a real project before choosing a paid plan; see current plans at rendero.ai/#pricing.
Don't see a term? Check the FAQ for workflow and pricing questions, or reach out at info@rendero.ai.