Courses Architectural Design Generative Architecture with AI 2.0 (SD+FLUX)

Generative Architecture with AI 2.0 (SD+FLUX)

Name: Generative Architecture with AI 2.0 (SD+FLUX)
Price: 119.00 EUR
Rating: 5 (4 reviews)

This workshop explores how to integrate Stable Diffusion, FLUX, and text-to-video tools into advanced architectural design workflows.

00:00

Overview
Content
Instructors
Reviews

Aug 2, 3, 2025

10 Lessons

Intermediate

8 Hours

Certificate — Learn

English

Unlimited Access

€140.00

€119.00

Gift Course

Introduction of the workshop:

The new Generative Architecture AI 2.0 (SD + FLUX + VIDEO) studio is a three-part sequel to Fredy Fortich’s PAACADEMY workshop. It retains the robust image-generation workflow of the first edition while introducing a complete, text-to-video pipeline, deeper FLUX 1.1 techniques, and next-gen SDXL control workflows.

By the end, architects will be able to transition seamlessly from 3D massing models to cinematic pitch videos. This workshop explores how Fredy is integrating generative AI into architectural practice, not as a replacement for critical thinking, but as a powerful extension of it.

This workshop will teach you how to refine AI workflows, integrate multimodal approaches, and elevate your design process with precision and creativity.

For detailed installation guides, please check the workshop’s Google Drive. To access it, click "Watch Now" → "Files" → "Download Course Files"

What You’ll Learn?

Overview of Generative AI applications being researched and developed at one of the top innovative architectural firms.

Weighted Prompting, LoRas, Seeds, Steps, ControlNet, and Denoising.

Customized SDXL Segmentation technique for Photoshop and/or Rhino.

Flux Upscaling Techniques.

Multimodal workflow: Rendering tools + Forge SD + Photoshop AI + Comfy SD/FLUX.

Advanced upscaling techniques for architectural workflows, including the use of LLMs + Diffusion models.

FLUX inpainting techniques.

FLUX Photoshop-styled Compositor custom workflow.

Img-to-Video workflows using Runway.

Generative Architecture with AI 2.0 (SD+FLUX)

Methodology:

As generative AI tools become increasingly integral to design workflows, the challenge is no longer just about using them. It’s about mastering them with intent. While simple prompts can spark ideas, they fall short when delivering site-specific, context-aware solutions.

As projects progress, architects need more than just automation; they require structured, deliberate methodologies that provide control over AI-driven processes.

In this workshop, you will learn how to work with a multimodal approach, utilizing your 3D model with traditional tools such as Rendering tools and Adobe Photoshop, as well as the latest technologies, including Stable Diffusion (SD), FLUX, and text-to-video models.

Fundamentals Lecture: AI in practice; Diffusion model basics, Workshop SDXL, segmentation primer
Image Mastery Workshop: ComfyUI build; LoRA applications upscaling & denoise techniques
Storytelling Workshop: AnimateDiff SDXL + motion-LoRA; Runway Gen-2 image-to-video

In this workshop, you will learn to use custom workflows using the basics you learned the day before. Next, you will learn Upscaling and a multimodal SDXL+Flux workflow that also leverages LLM advantages to improve your image precision and prompting efficiency.

You will receive multiple custom ComfyUI workflows, including Flux Denoising, Step Comparer, Compositing, Inpainting, and Upscaling, that will elevate your image refinement to the next level.

Finally, this workshop will teach you how to leverage all that you have learned to make a final video that
showcases all techniques: from segmentation to illustration to final video renders.

Generative Architecture with AI 2.0 equips you with exactly that speed-to-vision advantage, pairing Fredy Fortich’s rigorously tested SD + FLUX workflows, utilizing state-of-the-art video tools, to ensure your proposals stand out in competitions, pitches, and social media teasers alike.

What You’ll Use?

Highlights:

Content

Intro + PAACADEMY Updates (04:50)

Data-Driven Architecture (40:12)

AI-Powered Design at MVRDV

This session introduced MVRDV's integration of AI into architectural workflows, emphasizing data-driven design, sustainable practices, and the transformative potential of generative tools like Stable Diffusion, ComfyUI.

Architectural AI Customization Tools (39:25)

AI Rendering & LoRAs

This session explores how AI tools like ControlNet, ComfyUI, and custom-trained LoRAs can be used to render, upscale, and stylize architectural images with increased precision, allowing for rapid iteration, and unique design languages.

Diffusion Basics & Prompting Techniques (01:02:18)

Interfaces & Prompting

This session explores user-friendly tools built on Stable Diffusion, covering key settings that affect image quality and consistency, while introducing advanced methods such as prompt structuring, control systems, and model customization

ControlNet in Stable Diffusion (41:58)

Mastering Stable Diffusion

This session introduces Stable Diffusion techniques focusing on denoising strength, sampler selection, and ControlNet applications to guide users in achieving consistent, geometry preserving architectural renders

Segmentation in Generative Design (37:26)

Photoshop-Aided ControlNet Workflow

This session demonstrates how to guide architectural AI renders using Photoshop-style segmentation maps, denoising strength, and ControlNet weighting in Forge.

Architectural ControlNet Design Guide (56:27)

ControlNet Canny Workflow for Architecture

This session focuses on guiding architectural image generation using ControlNet, Canny edges, denoising strength, and seed control in ComfyUI.

Advanced ComfyUI Image Generation (57:34)

Multi-Control Workflow for Architects

This session demonstrates how to combine multiple ControlNets Canny, Depth, and IPAdapter within ComfyUI to generate precise architectural images.

Clip Prompting in Flux (56:20)

Flux Prompting Models

This session explores how Clip-based prompting in Flux workflows enables more expressive image generation, enhanced upscaling, and scene-aware modifications using ControlNet and Context models.

Targeted Image Editing Workflow (56:31)

Flux Context Inpainting & Control

This session explored Flux Context and ComfyUI workflows for selective image generation, inpainting techniques, image cropping for faster processing, and custom node installations for better control and efficiency in AI-assisted image editing.

Screenshots:

Instructors:

Fredy Fortich

Biography

Fredy Fortich is an Architect and Engineer specializing in computational design, with a particular focus on BIM coordination in Revit, performance-based design, generative design, and machine-learning methods. He holds a Master of Science in Building Technology from TU Delft in the Netherlands and a Master of Architecture from Universidad de los Andes in Bogotá, Colombia. Currently, Fredy works as a Technical Architect at MVRDV, BIM Manager at French Studio, and an AI researcher on Diffusion Models.

560 views

4 reviews.

Amir Daneshfar 2025-08-07 14:32

Hi, Thank you so much for the great workshop—it was incredibly insightful and covered a very interesting topic. Fredy was extremely knowledgeable and kind, and I really appreciated how he took the time to answer all the questions. Just one suggestion: I believe workshops like this could be even more engaging if students had the option to join live and interact directly, rather than only participating through the group chat. A live exchange would not only enhance the experience for the students but also give the instructor some real-time energy and feedback, making the session even more dynamic for everyone involved. Overall, this was a very valuable experience, and I’m grateful for the opportunity to take part. Thanks again! Best regards, Amir Daneshfar,

Kaloyan Valkanov 2025-08-21 22:52

Great workshop - a lot of useful info

Mertkan Günaslan 2025-08-24 19:47

It was a nice course ! Thank you!

Jean-Marc EMY 2025-09-14 20:57

Maybe the first part was a bit redundant compared to some previous courses, but I really enjoyed the ComfyUI part. The Lora and models that are not shared make it a bit complicated to follow the lesson but despite all this, it makes you want to dive deeper into these tools.