- Hyrise AI
- Posts
- 🤖 Cerebras and Core42 Enter Phase 2 for the World's Largest AI Supercomputers
🤖 Cerebras and Core42 Enter Phase 2 for the World's Largest AI Supercomputers
PLUS: AI-Driven Solution Emerges to Tackle California's Housing Crisis
Hyrise AI
November 14, 2023
Welcome, AI Enthusiasts.
Cerebras Systems and Core42 have entered the second phase of their project, Condor Galaxy 2, to build the world's largest interconnected AI supercomputers.
Holliday Development's affordable housing project, "The Phoenix," in California's San Francisco Bay area is leveraging AI to cut costs and streamline construction.
In today’s issue:
🤖 Cerebras and Core42 Enter Phase 2 for the World's Largest AI Supercomputers
🏠 AI-Driven Solution Emerges to Tackle California's Housing Crisis
🛠️ 3 New AI tools
💻 Custom prompts ChatGPT and DALL-E 3
🤖 3 Quick AI updates
Read time: 4 minutes.
LATEST HIGHLIGHTS
Image source: Cerebras
To recap: Cerebras Systems and Core42 have entered the second phase of their project, Condor Galaxy 2, to build the world's largest interconnected AI supercomputers. The network comprises nine interconnected supercomputers aiming to achieve a remarkable 36 exaFLOPs of AI compute capacity. This ambitious project reflects the commitment of Cerebras and Core42 to creating a constellation of AI supercomputers with unparalleled compute power.
The details:
Condor Galaxy 2 is a joint project between Cerebras Systems and Core42, aiming to create the world's largest interconnected AI supercomputers.
The network will consist of nine interconnected supercomputers and is projected to achieve a total of 36 exaFLOPs of AI compute capacity upon completion.
Cerebras Systems uses entire silicon wafers to print its cores, with each wafer having the equivalent of hundreds of chips. This approach allows them to incorporate 54 million AI-optimized compute cores in a single supercomputer.
Here is the key takeaway: Cerebras Systems and Core42 are collaborating on the construction of an ambitious network of interconnected AI supercomputers known as Condor Galaxy 2. This project aims to achieve a staggering 36 exaFLOPs of AI compute capacity, demonstrating their commitment to advancing AI research and applications across various domains, including healthcare, energy, climate studies, and more.
Image source: DALL-E 3
In Summary: Holliday Development's affordable housing project, "The Phoenix," in California's San Francisco Bay area is leveraging AI to cut costs and streamline construction. Using AI tools from Autodesk, the project's developers optimized the design to minimize noise from the nearby highway, maximize views, reduce construction waste, and cut costs. The technology has enabled the project to access subsidies worth over $25 million and reduce construction costs by around 25%. AI-driven design and construction innovations like these could help address California's housing crisis.
Key points:
AI-driven design and construction innovations are playing a significant role in addressing California's housing crisis, particularly by reducing costs and construction time.
Holliday Development's "The Phoenix" affordable housing project in West Oakland leveraged AI tools from Autodesk to optimize design, reduce noise pollution, maximize views, and minimize construction waste.
The project now costs about $400 per square foot, which is approximately 25% less than comparable buildings, making it financially viable. This was achieved by using AI to streamline the design and construction process.
AI technology has the potential to revolutionize the construction industry by optimizing material choices, reducing waste, and accelerating project timelines, offering a roadmap for other developers to follow in tackling housing affordability and construction challenges.
Our thoughts: AI-driven innovations in construction, such as those seen in "The Phoenix" project, hold great promise for addressing housing crises by significantly reducing costs and time, optimizing designs, and minimizing waste. These advancements mark a crucial step forward in the construction industry, offering a blueprint for more efficient and affordable housing solutions, especially in regions like California, where housing is in high demand and short supply.
TRENDING TECHS
💼 theGist- A unified workspace powered by AI that creates a radical personalized focus getting you the gist of your work.
🌎 WebWave-Create a website like no other with AI
🗂 SELLDONE- The next-generation of business operating systems
AI DOJO
ChatGPT
Innovative Business Model:
Prompt: "Develop a business model for a company that uses drones to plant trees in deforested areas."
DALL-E 3
Imaginary Technology:
Prompt: "Illustrate a futuristic kitchen gadget that doesn't exist yet, combining high-tech and vintage design elements."
QUICK BYTES
Meta's director of engineering for Generative AI, Sergey Edunov, estimated that only two new nuclear power plants would be needed to meet the growing demand for AI applications for the next year, particularly for AI inference. He explained that Nvidia's release of one to two million of its H100 GPUs next year, used for AI applications, could be powered by just two nuclear reactors. However, this estimation focused on the inference side, while the challenge for training large language models (LLMs) lies in acquiring enough data, which could require tapping into alternative sources, such as video data, to scale higher-performing models.
Organizations are increasingly adopting multi-cloud strategies, and in a survey, 95% of respondents expressed intentions to use a multi-cloud approach. However, cloud waste is a significant issue, with an average of 40% of cloud spending going to unused applications and storage annually. To address this, organizations can focus on building a more efficient multi-cloud ecosystem, utilizing smarter managed services, increasing cost transparency, simplifying pricing models, fostering greater multi-cloud collaboration, finding customized cloud services, and setting a shared vision with partners and customers to optimize their cloud spending and operations.
Silo AI, a European AI lab and services company, has introduced Poro, an open-source language model designed to cater to European languages and encourage AI research and development in the region. Poro aims to fill the gap in language models available for European languages, which are often underrepresented compared to English-centric models. This initiative by Silo AI seeks to advance AI capabilities and applications in Europe while fostering collaboration and innovation within the AI community.
SPONSOR US
🦾 Get your product in front of AI enthusiasts
THAT’S A WRAP