The centerpiece announcement was Vera Rubin, first teased at Computex 2024 and now scheduled for release in the second half ...

At Nvidia's GTC conference this week, it showed off two new AI chip designs: Vera Rubin and Feynman. These AI-first processors are designed to be the future of AI and robotics in 2027 and 2028.

NVIDIA has now confirmed that its next-gen AI GPU will be codenamed Feynman, after renowned American theoretical physicist, Richard Phillips Feynman, who is known for his work on quantum mechanics ...

It's astonishing that they can accomplish anything with a crummy format like FP4. I mean that seriously - it can only represent 16 values with a severely limited dynamic range. It's a tribute to ...

In addition, the company decided to announce its roadmap for Rubin Next — to be introduced in 2027 and Feynman AI chips in 2028. The major innovation for NVDA’s new chipsets is a qualitative ...

But I'll have the best laugh of the decade if after all this effort, things get worse for AI's results instead of better. I'll still have a really good chuckle for the month if any improvement is ...

A single Ultra chip will offer the same 20 petaflops of AI performance as Blackwell, but with 288GB of memory — up from 192GB in vanilla Blackwell. On the far horizon are Feynman GPUs.

In addition, the company decided to announce its roadmap for Rubin Next — to be introduced in 2027 and Feynman AI chips in 2028. The major innovation for NVDA’s new chipsets is a qualitative shift ...