NVIDIA Unveils Blackwell Extremely GB300 Amped With 288GB Of HBM3e

NVIDIA CEO Jensen Huang introduced Blackwell Extremely right now on stage at GTC, the place he described it as NVIDIA’s strongest GPU so far. It actually suits that description: NVIDIA says Blackwell Extremely presents 15 petaflops of dense FP4 tensor compute, a 50% uplift from Blackwell GB200. The actual story is the RAM, although: 288GB of HBM3e reminiscence on a single package deal.

Meaning a rack of Blackwell Extremely GB300 Superchips, equivalent to within the NVIDIA GB300 NVL72, presents at least 20TB of HBM reminiscence and a staggering 1.1 exaFLOPS of FP4 compute because of the 72 Blackwell Extremely GB300 chips on the 36 ‘Superchips’ inside. As earlier than, every pair of Blackwell Extremely GPUs is paired with a Grace CPU and its provide of LPDDR5X reminiscence, in addition to some 130 TB/s of NVLink connectivity between superchips and 14.4 TB/sec of off-rack networking.

In fact, as earlier than, the “level” of the NVL72 racks is that they function as a single huge “GPU”. Certainly, Huang refers back to the GB300 NVL72 as “the final word scale-up,” for which he says there may be “no alternative.” Past the spec upgrades, Blackwell Extremely is outwardly additionally coming together with new directions to speed up consideration in AI computations that NVIDIA says end in double the efficiency for that particular workload.

Jensen’s on-stage presentation was closely targeted on defining datacenters as “AI factories,” to the purpose that Huang claims that each one “companies with factories” will quickly have to have “two factories”: one for manufacturing, and one for AI manufacturing, to make the AI that powers the merchandise. He additional claimed on stage that Blackwell presents 40 occasions the “token income” versus Hopper when configured this manner.

Blackwell Extremely is coming within the second half of 2025, within the type of NVL72 racks in addition to HGX B300 NVL16 methods with sixteen GPUs inside. In fact, these NVL72 racks may also be organized into DGX SuperPODs with as many Blackwell Extremely GPUs as you’ll be able to afford. As Jensen says, “the extra you purchase, the extra you save.” For cloud service suppliers internet hosting AI companies, which will really be the case.