In this video, I test Rendair, an AI-powered platform that takes you from a simple SketchUp model to photorealistic architectural renderings—all inside one collaborative workspace.
Instead of switching between Photoshop, PowerPoint, Miro, and multiple AI tools, you can generate, edit, compare, and organize your design iterations in one place.
In this video you'll learn: AI rendering from SketchUp 3D models Convert simple massing models into photorealistic architecture Test different facade materials with AI Replace materials using texture references AI-powered image editing Replace plants with specific species Add and edit people in renderings Generate architectural detail close-up renderings AI Floor Plan to Render workflow Create immersive 3D interior scenes from images Organize design options in a collaborative workspace Upscale renderings to ultra-high resolution Share projects with clients and design teams
Whether you're an architect, landscape architect, urban designer, interior designer, or architecture student, this workflow can dramatically speed up your visualization process while keeping your design iterations organized.
This video introduces an AI-powered architectural visualization tool that provides a comprehensive workflow from 3D models to high-resolution renderings. The tool features a collaborative, Miro-like workspace where users can manage facade variations, seasonal changes, and detailed edits in one place, significantly streamlining the design iteration and client presentation process.
I finally find an AI tool that provides a complete workflow for architecture visualization.
It takes you all the way from 3D model to high quality renderings with accurate image editing,
facade variations, seasonal changes, and ultra-high-resolution detail renderings.
All within one collaborative workspace, similar to Miro.
You can keep tracking of every rendering, compare facade design options,
edit image, and even generate floor plan renderings all in one place.
So, let's get started.
We'll begin with the 3D model.
Here, I'm using a SketchUp model of a villa that I designed for a client.
Export your model in one of the supported 3D file formats.
Then, open Render.
You'll find the link in the description below.
In the App section, choose 3D model to render, and simply drag and drop your model into the workspace.
Unlike traditional rendering workflows, you don't need to model every detail.
A simple massive model is usually enough for AI rendering.
You can describe the architectural details in the prompt, such as facade materials,
windows, lighting, landscape, and the surrounding environment.
And let the AI render the design based on your instructions.
Here are some pre-made rendering styles, including realistic sketch, watercolor, marker, and more.
Each style comes with example images.
So, you don't need to train your own model, just pick the style you like.
For this example, I'll choose realistic when rendering 3D models.
I usually prefer using ChatGPT2 or NanoBanana, as they produce very consistent results.
Then, simply choose how many images you'd like to generate.
I'll start with one image.
The AI generates the first rendering, and it matches my prompt really well.
From the architectural details to the surrounding environment.
But I'd like to add a few more details.
So I go to edit from the top menu, draw a selection around the area I want to change,
and simply type a prompt like add a black car here.
A few seconds later, the AI updates only the selected area,
while keeping the rest of the rendering unchanged.
Using the same workflow, I'll render the back of the building,
and use the previous rendering as a style reference to keep everything consistent.
Here's the backyard rendering I generated.
Next, let's start testing different facade materials.
I'll create a new board, drag in the renderings we just generated.
Click on one of the renderings, then go to edit and select change materials.
In the prompt, I simply describe the change.
Change the second floor facade to timber.
A few seconds later, the AI updates only the second floor facade,
while keeping the rest of the building exactly the same.
We can also provide a reference texture image.
For example, I'll use this brick texture as the reference,
and the AI will accurately apply it to the new rendering
while preserving the building geometry and the lighting.
In the workspace, I can easily organize and compare different facade options.
Without downloading images or arranging them in PowerPoint,
everything stays in one place,
making it easy to review design iterations,
compare materials, and share the board directly with your team or clients for feedback.
Another feature I find really helpful is upscale.
Just click the dot on the right side of the image,
scroll down, and select upscale.
The AI will generate an extremely high resolution version of the rendering.
Allowing you to zoom in and inspect all the fine details.
This upscaled image is around 32 MB,
making it perfect for large format presentations, printing, or client deliverables.
We can also use the edit function directly in the workspace.
For example, I'll select the car,
then enter a prompt like reverse the car.
Since it's a residential house,
having a car speeding toward the entrance feels a bit unrealistic.
Reversing the car makes the scene look much more natural and believable.
Let's keep editing.
The plant replacement feature is especially useful.
I can mark the exact areas I want to update
and ask the AI to replace the existing plants with specific species.
It places the new plants accurately and naturally
while keeping the building and the rest of the scene exactly the same.
Another really useful feature in the editing canvas is the close-up view.
Just select the area you want to focus on
and enter a prompt like generate a close-up view.
The AI will create a detailed rendering of that selected area
while keeping the same materials, design, and visual style.
I can keep selecting different areas and generating close-up views.
By the end, I have a complete series of high-quality architectural detail renderings,
all organized in the workspace for easy review and comparison.
I can also add people to the renderings.
Just choose populate render, then describe the people you want.
In this example, I asked for one man and a dog.
Here are the two rendering options the AI generated.
I can further refine the people in the rendering by clicking fix people.
From there, I can adjust attributes like gender,
age, and ethnicity to better match the building's intended users.
This is a feature I use quite often
for the final touch of architectural renderings in real-world projects.
For small architectural projects,
this is definitely a feature you don't want to miss.
Floor plan to render.
Simply drag your floor plan into the workspace,
then choose the floor plan to render function and enter your prompt.
There are plenty of pre-made styles to choose from,
including 3D, realistic blueprint, and more.
I tried all of them, and they each produce a different look.
Let me know in the comments which style do you use most often in your projects.
Last, here's a really cool feature, image to scene.
It creates an immersive 3D experience for your client,
almost like a virtual property tour.
First, I export a basic interior view from SketchUp,
and use image to render to turn it into a realistic interior rendering.
Then, I go back to the app section,
select image to scene, and drag the finished rendering into the workspace.
You can add a prompt with more detailed instructions,
or simply leave it blank, then click generate.
In just a few moments, you'll get an immersive 3D interior scene
that clients can explore interactively.
It's a much more engaging way to present your design
than a traditional static rendering.
All of the generated images, videos, and project steps
can be kept together in the same workspace.
You can share the board with your team or clients
for markups, comments, reviews, and presentations.
Overall, it creates a really smooth AI rendering workflow
for the entire project development process.
Try Render using the link in the description,
and let me know how it works for you.
I'd love to hear about your experience with it.