Plus, how solar fits into the next 10 years| www.dwarkesh.com
We love TPUs at Google, but GPUs are great too. This chapter takes a deep dive into the world of NVIDIA GPUs – how each chip works, how they’re networked together, and what that means for LLMs, especially compared to TPUs. This section builds on Chapter 2 and Chapter 5, so you are encouraged to read them first.| jax-ml.github.io
NVIDIA today announced it is working with Google Quantum AI to accelerate the design of its next-generation quantum computing devices using simulations powered by the NVIDIA CUDA-Q™ platform.| NVIDIA Newsroom
Quantum dynamics describes how complex quantum systems evolve in time and interact with their surroundings. Simulating quantum dynamics is extremely difficult yet critical for understanding and…| NVIDIA Technical Blog
Large language models (LLMs) offer incredible new capabilities, expanding the frontier of what is possible with AI. However, their large size and unique execution characteristics can make them…| NVIDIA Technical Blog
To get the most out of AI, optimizations are critical. When developers think about optimizing AI models for inference, model compression techniques—such as quantization, distillation…| NVIDIA Technical Blog
“China wants it all, and they can have it"| www.dwarkesh.com
Fang-Pen Lin's blog about programming| Fang-Pen's coding note
Scaling reinforcement learning, tracing circuits, and the path to fully autonomous agents| www.dwarkesh.com
This Gradient Updates issue explores how much energy ChatGPT uses per query, revealing it’s 10x less than common estimates.| Epoch AI
Transformer Engine, part of the new Hopper architecture, will significantly speed up AI performance and capabilities, and help train large models within days or hours.| NVIDIA Blog
Of scalpels and Swiss cheese. “The dumbest thing you could do is compromise and build your house in the middle of the street.”| www.chinatalk.media
Fugatto generates or transforms any mix of music, voices and sounds described with prompts using any combination of text and audio files.| NVIDIA Blog
To enable a society that co-exists with AI, SoftBank Corp. (TOKYO: 9434) is continuing to make progress on its goal to build and provide different types of AI infrastructure in Japan. On October 31, SoftBank announced it finished installing approximately 4,000 NVIDIA Hopper GPUs in its Japan top-lev…| SoftBank News. Bringing IT closer to home.
Discover how to get Mistral’s new multimodal LLM, Pixtral 12B up and running on an Ori cloud GPU.| blog.ori.co
---| Google Cloud
Just one year after its launch, ChatGPT had more than 100M weekly users. In order to meet this explosive demand, the team at OpenAI had to overcome several scaling challenges. An exclusive deepdive.| newsletter.pragmaticengineer.com
Listen now | “They're confusing skill and intelligence.”| www.dwarkeshpatel.com
For the last several decades, one avenue of technological progress has towered over nearly everything else: semiconductors.| www.construction-physics.com
Today brought some electrifying news from Mark Zuckerberg, the CEO of Meta, who has announced the ongoing training of LLAMA 3| Llama LLM
Bring accelerated computing into any data center with modular server designs.| NVIDIA
Microsoft unveils two custom chips, new industry partnerships and a systems approach to Azure hardware optimized for internal and customer workloads| Source
(generated by Midjourney)| eastwind.substack.com
In short The generative AI market surpassed $25.6 billion in 2024, driven by rapid adoption and the increasing integration of AI capabilities across industries, according to the 263-page Generative AI Market Report 2025–2030 (published January 2025). The data center GPU market saw remarkable growth to $125 billion, with NVIDIA maintaining a dominant position, holding 92% of the market share. Microsoft and AWS lead the rapidly expanding foundation models and model management platforms marke...| IoT Analytics
Designed to Deliver Revolutionary Performance!!| NVIDIA
Transformer models apply an evolving set of mathematical techniques, called attention or self-attention, to detect subtle ways even distant data elements in a series influence and depend on each other.| NVIDIA Blog
Note: I was saving this for the launch of the Symmetry Institute, but given the recent discussions around REBUS/CANAL, Deep CANALs, and Neural Annealing I pushed it forward.| Opentheory.net
Where is AI Going Over the Next Two Decades? The Sci-Fi Story that AI Will Destroy All the Jobs Will Slowly Fade as AI Becomes More Deeply Collaborative and Co-Creative.| danieljeffries.substack.com
