In partnership with

Welcome to this week’s edition of Architecture Insights. A weekly newsletter on artificial intelligence for architects, landscape architects, and designers.

Mark Zuckerberg and Meta release their most capable model to compete with ChatGPT 4 and Claude 3. Adobe announces an AI image enhancer that improves image quality by up to 8x along with more updates to their most popular AI tool, Firefly.

News & Updates

1. Meta & Llama 3

Meta recently released two new versions of Llama 3 large language model (LLM) - Llama 3 8B and Llama 3 70B. Here's why these updates are relevant for architects and design professionals:

Multimodal Capabilities. One of the key planned updates for Llama 3 is multimodal processing, allowing it to understand and generate both text and images.

Improved Language Understanding. The Llama 3 models were trained on a much larger dataset (15 trillion tokens) compared to Llama 2, including a significant amount of code and multilingual data.

Longer Context Window. Meta is working on increasing the context window size for Llama 3, allowing it to process and understand longer pieces of text 2. This could be beneficial for architects and designers working with lengthy design briefs, technical reports, or legal documents, enabling the AI to provide more accurate summaries and insights.

Domain-Specific Expertise. Meta plans to develop specialized versions of Llama 3 that are experts in specific domains like engineering and architecture 1. These domain-specific models could potentially provide more accurate and relevant assistance for tasks related to architectural design, construction, and project management.

2. VideoGigaCAN by Adobe

VideoGigaGAN Upscaling
Adobe has announced a new video upscaling model called VideoGigaGAN that can upsample a video up to 8x while adding rich details.

The model can take low-resolution, blurry video input and produce high-quality, sharp output with impressive detail, even in areas like faces, skin, hair, and water movement.

This technology could be highly relevant for architects and landscape architects, as it allows them to take low-quality footage or renderings of their designs and upscale them to create stunning, photorealistic visuals.

3. Adobe Firefly 3

It seems like almost every month (in this case week) that Adobe releases new updates to Firefly and their suite of AI tools. We have already explored their "Structure Reference" and others recently released but here are the key takeaways from their recent update for Firefly 3: 

Better photorealism and detail. Generates more photo-realistic images, especially for scenes featuring people.

Enhanced creative control. New "Structure Reference" and "Style Reference" features allow users to guide the composition and aesthetic of generated images using reference images. Expanded range of output styles including illustrations, photographic art, and vector art

Improved prompt understanding. Better ability to accurately reflect longer and more complex text prompts. Enhanced quality of text rendering within generated images.

Increased productivity. Features like "Generative Expand" that can automatically fill and resize images.

Resources

If you have ever considered building your own custom AI that gets trained on your project and company data then AE Studio is the software to explore those possibilities.

Hire a world class AI team

Engineers who understand AI are expensive and difficult to find, and it can be hard to figure out who to trust. On top of that, 85% of all AI projects fail.

But AE Studio succeeds.

We listen to your business challenge and help you craft and implement the optimal AI solution with our team of world class AI experts from Harvard, Stanford and Princeton.

Our development, design, and data science teams work closely with founders and executives to create custom software and AI solutions that get the job done. The secret to our success is treating your project as if it were our own startup.

Tell us all about your big AI Idea

Resume Critiquing with AI & LinkedIn

Upload an image of your resume or cover letter to ChatGPT-4 or an alternative version just as capable i.e Claude 3, Gemini 1.5, etc…

Start your prompt with the following:

“I am applying for a position as a (Position title) at (name of firm). I have attached my resume and would like you to review my credentials and offer a critique of it and let me know which areas stand out as good and which areas I might consider changing based on the below job description taken from LinkedIn.”

*Attach your resume (this is where you require GPT-4 or a similar alternative if you want to submit PDFs, otherwise you will have to submit your resume in text format).

*Copy and paste the job description from LinkedIn, here is an example:

Include everything starting from the ‘About the job’ section until the end of the job description.

Try this the first time around and evaluate the response from ChatGPT and then repeat the same process if you want to fine-tune your answer.

Adding more to your prompt like “focus on the technical skills over personal achievements”. This can tell your large language model which areas to give more attention to and improve on.

AI Image of the week

Thank you for reading this week’s issue, check past issues here. Share this newsletter with colleagues, friends, or anyone interested in the combined world of architecture and artificial intelligence.

Until next Friday,

A.I.