Architects' tools
1.16K subscribers
375 photos
37 videos
62 files
69 links
Hi, my name is Albert Sumin, this is my channel, I am an architect and computational designer at Cloud Cooperation (Vienna, Austria)
Download Telegram
Renders from last posts are done with Google Gemini 2.5 Flash Image inside #comfyui . Another example of what #gemini does well is these kinds of scenes. Such views are especially useful if we want to make an animation, and it is also sometimes necessary to show a particular detail of a building up close.
🔥16😁3👏1
I'm almost done bothering you with this project. This time, it's a quite common task of changing the atmosphere of an image.
🔥121👍1👏1
New free workshop announcement to be held on the Design Morphine platform at the end of January. I talk a lot about AI rendering here, but in this workshop we will try the opposite approach, where we first have a concept created with AI, and based on this concept we need to create a complex 3D model using standard tools such as Grasshopper and Blender. You can sign up now. Difficulty level: medium (or even higher, since we only have two days for a complex task, which means we will have to skip many basic things in the explanations).

https://designmorphine.com/education/emergent-futures-v1-0

#education #grasshopper #blender #ai
7🔥1
Media is too big
VIEW IN TELEGRAM
A video related to the project I posted last week. #gemini #veo #flow
🔥7
This media is not supported in your browser
VIEW IN TELEGRAM
Another experiment with Gemini 2.5 Flash Image. All views are based on a single screenshot from Google Earth + a reference with architectural style. This process can be used when you need to prepare an idea in just a few hours, without having any 3D models or drawings for the project. #comfyui #gemini
🔥151
There is a lot of interesting news on AI today. I don't usually publish this kind of thing, as you can read about new models everywhere without me, but let's make an exception.

- First, of course, the new gemini-3-pro-preview (nano banana pro). I'll show some my own tests in the next post, but here is the page with the official announcement:
https://deepmind.google/models/gemini-image/pro/

- Second, Meta has released a free, open-source model for 3D generation and image and model segmentation. I haven't tried it yet, but the preview is impressive:
https://ai.meta.com/sam3d/

- Open AI has created group chats for users. It's not particularly useful, but it's fun:
https://openai.com/index/group-chats-in-chatgpt/

#ai
🔥1👏1
So, the new Gemini-3-Pro-Preview, a few points:

- The level of details is like nothing we've seen before; for some shots it doesn't look like AI at all. After the initial generation, the model can upscale the image without losing details to 2k or 4k, depending on the settings.
- It is not very stable yet; in fact, you can only try it through the API, and not all generations are successful there. I was unable to generate an image in 4k, I managed to do it once, but I couldn't save it, the site crashed, and I only got a glimpse of the result.
- AI rendering is far from ideal in terms of preserving the composition of the original image. Every second generation creates a completely different view related to input, but maybe this is a problem with Google AI Studio, which confuses the context with previous requests. We'll see tomorrow in other services.
- The camera change and image editing works better than in the previous version.
- The price has increased fourfold, but maybe the stable version (not the preview) will work four times more reliably, meaning fewer generations will be needed to achieve the result. There are reasons to believe this.
#gemini
4🔥2
878afaed-8312-4ada-befb-1394d847af0d.jpg
3.2 MB
Here is an uncompressed 2k image so you can see the details.
🔥21