Meta has purchased NVIDIA's new Blackwell B200 AI GPUs to train its Llama models, according to Meta CEO Mark Zuckerberg. The company is also training a third-generation of its Llama model on two GPU clusters that it announced last week, each of them packing around 24,000 of NVIDIA's current-gen Hopper H100 AI GPUs.
The news is coming from a new report by Reuters, which said that Meta will continue using its current H100-powered AI GPU clusters to train its current-gen Llama 3 model, but will use NVIDIA's new Blackwell B200 AI GPUs to train future generations of the model, according to a Meta spokesperson.
Popular Now: GTA 3, Vice City, and Liberty City Stories are coming to the New Nintendo 3DSNVIDIA announced its new Blackwell B200 AI GPU at its GPU Technology Conference (GTC) event this week, offering gigantic improvements to all things AI.
Meta wants the best of the best, which is where the orders from the social networking giant for the best AI GPUs ever made come into play to power future Llama models. Meta has been one of NVIDIA's biggest customers, buying up big quantities of AI GPUs over the last couple of years and that isn't slowing down as now Blackwell is nearly here.
- Read more: NVIDIA CEO: our next-gen Blackwell GPU will cost $30,000 to $40,000 each
- Read more: NVIDIA spent $10 billion on developing its next-generation Blackwell GPU
- Read more: NVIDIA GB200 Grace Blackwell Superchip: 864GB HBM3E, 16TB/sec bandwidth
- Read more: NVIDIA Blackwell AI GPU: multi-chip die, 208B transistors, 192GB HBM3E
- Read more: NVIDIA's new Blackwell-based DGX SuperPOD: ready for trillion-parameter scale for generative AI
I recently reported that Meta CEO Mark Zuckerberg had a long-term vision for AGI (artificial general intelligence) that would be powered by the equivalent of 600,000 x NVIDIA H100 AI GPUs. Now that Blackwell is ramping up to release, Meta can super-charge its AGI efforts much faster.
