NewCrit Space is open to everyone. Invite people to your project.Open a project

How to Render a Rhino Model with AI, Step by Step

Shaun McCallum

Shaun McCallum

October 20, 2026

One timber house in the Scottish Highlands, from a plain .3dm to a daylight render, a dusk version, a set of detail shots and a short video.


To render a Rhino model with AI in Fenestra, drag the .3dm file onto a Crit Space board, orbit to the view you want, set the sun and time of day, then connect that view to an Image Generator with a short prompt for materials and setting. The model sets the form and camera. The prompt sets everything around it.

The video above runs the render in a few minutes. Below are the same steps for one project, a black timber house in the Scottish Highlands, followed by what we made from that first render: a dusk version, two sets of detail shots and a short video.

Photorealistic render of a black timber house with a glazed corner on a heather hillside in the Scottish Highlands, with birch trees and low mist
The finished render, made from a plain Rhino model

What do you need before you start?

A Rhino model saved as .3dm, and a Fenestra account. Any plan works, including the free plan. You don't need to texture the model or set up lights in Rhino. Plain massing is enough, because the prompt does the materials.

It helps to tidy the model first. Hide construction lines, reference geometry and anything outside the view you want to show. What the camera sees is what the AI works from.

Step 1: How do you get the model into Fenestra?

Create a new project and choose Crit Space. In the preset picker, pick 3D to Render and click Use preset. The board opens with a 3D Model, a Text Prompt, an Image Generator and a Results node already connected.

The Crit Space preset picker with the 3D to Render preset selected
The 3D to Render preset sets up the whole chain for you

Drag your .3dm file onto the canvas and connect it in place of the sample model. The same works for .fbx and .glb files if part of the project lives in another tool.

Step 2: How do you frame the view?

Click the model to load it, then drag with the left mouse button to orbit. Frame it the way you would for a client image: eye level for a street view, a little higher to show the roof form. For this house we stayed near eye level, with the glazed corner turned towards the camera and trees on both sides to frame it.

The Rhino model, a text prompt and an Image Generator connected on the Crit Space canvas, with the render on the right
The model, prompt and generator on one board

Open Camera in the 3D Controls panel and switch to two-point perspective for exteriors. It keeps vertical lines straight, which is what makes an architectural image read as a photograph rather than a screenshot. Save the angle so you can come back to the same view after a design change.

Step 3: How do you set the light?

The light in the model view carries through to the render, so set it here rather than leaving it all to the prompt.

  • Time of Day moves the sun and the shadows. Late afternoon gives long shadows that show depth on the facade.
  • Sun Brightness sets how hard those shadows are.
  • Ambient Strength lifts the shadowed sides so they don't go black.
  • Ambient Rotation shifts reflections and highlights on glazing.
The 3D Controls panel with camera, sun and ambient light settings next to the model view
Camera and light controls sit next to the model

Step 4: What should the prompt say?

Describe what the model doesn't show: materials, landscape, location, weather. Leave the form alone, because the model already has it. For this house we used:

"Turn this 3D model into a photorealistic architectural render, natural daylight, realistic dark timber materials and wild landscaping. Atmospheric sky. Foggy Scottish Highlands in the distant background."

Name one main material and one setting. A prompt that lists five materials tends to spread them across the building in places you didn't intend.

Step 5: Which settings should you generate with?

The framed view plugs into the Image Generator as its input image. We used:

  • Model: Muse Image Edit
  • Aspect ratio: 16:9
  • Image count: 2
The Image Generator with the prompt and the 3D view as inputs, next to the finished render
The 3D view goes in as the input image

Two images per run gives you a choice without burying you in options. Some models take up to four reference images, so you can also connect a material sample or a photo of the site.

Step 6: How do you check the render?

Compare it against the model view before you share it. The massing, openings and camera should match. Look closely at window divisions, balustrades and anything thin or repeated, because that's where AI is most likely to improvise.

Before
After
Drag to compare the Rhino view with the render

Here the roof line, the glazed corner and the slot windows all hold. If something drifts, change the prompt and branch a new generator from the same view. The first result stays on the board, so you can compare the two.

How do you change the weather or time of day?

Drag out from the finished render and add a new Image Generator. The render becomes its input, so the building and camera stay put and only the conditions change. Keep the prompt to the change itself, such as dusk, warm light inside or heavy rain.

The black timber house in overcast daylight with mist in the valley
Day
The same view of the house at dusk with warm light behind the glazing
Dusk
Same view, branched from day to dusk with the interior lit

Each branch stays linked to the render it came from, so a client can see the options side by side and you can always get back to the original.

The black timber house at dusk in heavy rain, with a ginger cat walking through the heather in the foreground
A rainy dusk branch, with a visitor

How do you get detail shots from the same render?

Connect the render to a Multi Shot node. It makes nine new shots in one generation: aerials, close-ups of materials and junctions, and views from other sides. You don't describe each view. Add a short list of things you want to see, and it spreads them across the nine.

These came from the daylight render. The cladding and window frame close-ups are the kind of shot that would take a separate camera, materials pass and render in a traditional workflow.

An aerial shows the deck, steps and site.

Run Multi Shot again from the dusk branch and you get a second set in the evening light. Click Add all to board to place the nine shots as separate images, then keep the ones you want.

The dusk aerial, with the valley lights coming on.

Multi Shot invents what the model doesn't contain, like the hallway behind that door. Use these shots for mood and storytelling, and check anything a client might read as a design decision.

How do you turn the renders into video?

Add a Video Generator and connect two renders: one as the start frame and one as the last frame. Describe the change between them in a few words. This clip goes from the daylight render to the dusk aerial, so the camera rises as the light fades.

From the daylight render to the dusk aerial, in one clip

The video model fills in the frames between, so treat it as a presentation shot rather than a walkthrough of the model. The Image to Video guide covers the settings.

What does the finished board give you?

One Rhino model, one view and a short prompt became a daylight render, a dusk version, two sets of detail shots and a video. They all sit on one board, linked to the model they came from. Click Share in the top-right to invite your team, or create a view-only link for a client.

For more on turning Rhino models into photorealistic renders, including how it compares with a traditional render engine, see our Rhino page. The 3D to Render guide lists every control in the 3D panel, and the Multi Shot guide covers prompting for detail shots.

Render your Rhino model

Start a board

Frequently asked questions

Yes. Drag the .3dm file straight onto a Crit Space board. You orbit the model in the browser, and the framed view goes into the Image Generator as its input.

Crit Space accepts .3dm, .fbx and .glb files, so the same steps work for models from Rhino, SketchUp, Blender and most other 3D tools.

The render follows your model view for massing, openings and camera, and the prompt sets materials, landscape and atmosphere. Check fine details such as window divisions against the model before sending a render to a client.

Yes. Branch new generators from the first render to change the weather or time of day, use Multi Shot for nine detail shots and new angles, and animate between two renders with the Video Generator.