VerilogEval is an odd coding dataset - it gives the agent access to the golden testbench, which you will use to evaluate the final design from the model. I'm honestly surprised how papers like VerilogCoder or ChipAgents only get ~95% accuracy in this setting... That's what

normalcomputing.com

Wishing a happy Independence Day to all! One of my favorite holidays. 🇺🇸🇺🇸🇺🇸🎆🎆🎆

MAHA is focused on industrial chemicals like food dyes. But the molecules below are "naturally occurring" in today’s fruits and veggies. This is why MAHA may fail to solve chronic disease. To truly solve the epidemic, we must promote a meat-based diet.

Check out this bold new work from Normal Computing 🧠🌡️ — forecasting a golden age of computing driven by physics-based hardware and architectures!

Really excited to announce our new preprint, "Solving the compute crisis with physics-based ASICs!" For years, I’ve argued we are returning to the era of specialized hardware. We moved from CPUs to GPUs, then to domain-specific accelerators. We are now facing a clear compute

For this Perspective article, we brought together leaders in the field from industry, academia, and the public sector. It was a great collaboration and highlights the growing ecosystem around physics-based ASICs. We gave careful thought for how to define the term "physics-based

Here’s a thread on our new perspective article Solving the compute crisis with physics-based ASICs. To meet the demands of AI, a variety of new approaches to computing have been proposed that reimagine the basic building blocks of computers by considering the intersection of

A good friend of mine told me today that he’s been trying a steak-heavy (almost carnivore) diet, inspired by my pro-carnivore message. He said his energy has gone way up and so has his quality of life. I’m so happy to hear this - healing and helping people is my dream. If

Today at 4pm ET, Max Aifer will be presenting our Physics-Based ASICs paper to the Active Inference Institute. Check out the live stream 👇

Here's the livestream I did today with Active Inference Institute about physics-based chips for AI

Normal is growing! Welcome Erik, Cameron, Arvind, Prashant, and Hanna!

Today, the team at Normal Computing announces that we have achieved first silicon for CN101: the world’s first thermodynamic computing chip. CN101 accelerates AI inference, large-scale linear algebra, and stochastic sampling for diffusion models, marking a key step toward up to

Vision papers are fun to write, but they usually remain just visions. But every now and then a vision becomes real. Its a pleasure announce successful tapeout of the first thermo computing chip, realizing the vision in our article on physics-based ASICs arxiv.org/abs/2507.10463

Nice article in Semiconductor Engineering where Thomas Ahle, Doyun, and Hrolfur discuss the impact of Multi-modal AI on EDA tools.

Normal Computing has completed the tape out of CN101, the world’s first thermodynamic computing chip for AI/HPC data centers. Unlike traditional silicon, it uses thermodynamics, noise, and stochasticity to efficiently solve probabilistic problems like AI training, achieving up