Sundar Pichai on Gemini, AI progress and extra

Infrastructure for the AI period: Introducing Trillium

Coaching state-of-the-art fashions requires a variety of computing energy. Trade demand for ML compute has grown by an element of 1 million within the final six years. And yearly, it will increase tenfold.

Google was constructed for this. For 25 years, we’ve invested in world-class technical infrastructure. From the cutting-edge {hardware} that powers Search, to our customized tensor processing models that energy our AI advances.

Gemini was skilled and served fully on our fourth and fifth technology TPUs. And different main AI corporations, together with Anthropic, have skilled their fashions on TPUs as properly.

Right now, we’re excited to announce our sixth technology of TPUs, referred to as Trillium. Trillium is our most performant and most effective TPU up to now, delivering a 4.7x enchancment in compute efficiency per chip over the earlier technology, TPU v5e.

We’ll make Trillium out there to our Cloud clients in late 2024.

Alongside our TPUs, we’re proud to supply CPUs and GPUs to assist any workload. That features the brand new Axion processors we introduced final month, our first customized Arm-based CPU that delivers industry-leading efficiency and vitality effectivity.

We’re additionally proud to be one of many first Cloud suppliers to supply Nvidia’s cutting-edge Blackwell GPUs, out there in early 2025. We’re lucky to have a longstanding partnership with NVIDIA, and are excited to convey Blackwell’s breakthrough capabilities to our clients.

Chips are a foundational a part of our built-in end-to-end system. From performance-optimized {hardware} and open software program to versatile consumption fashions. This all comes collectively in our AI Hypercomputer, a groundbreaking supercomputer structure.

Companies and builders are utilizing it to deal with extra advanced challenges, with greater than twice the effectivity relative to only shopping for the uncooked {hardware} and chips. Our AI Hypercomputer developments are made doable partly due to our strategy to liquid cooling in our knowledge facilities.

We’ve been doing this for almost a decade, lengthy earlier than it grew to become state-of-the-art for the {industry}. And right now our whole deployed fleet capability for liquid cooling methods is almost 1 gigawatt and rising — that’s near 70 instances the capability of some other fleet.

Underlying that is the sheer scale of our community, which connects our infrastructure globally. Our community spans greater than 2 million miles of terrestrial and subsea fiber: over 10 instances (!) the attain of the following main cloud supplier.

We’ll hold making the investments essential to advance AI innovation and ship state-of-the-art capabilities.