Skip to playerSkip to main content
  • 1 day ago
Transcript
00:00And so today, we're announcing something very special.
00:04We're announcing NVIDIA NVLink Fusion.
00:10NVLink Fusion is so that you can build semi-custom AI infrastructure.
00:20Not just semi-custom chips, because those are the good old days.
00:26You want to build AI infrastructure.
00:28And everybody's AI infrastructure could be a little different.
00:32Some of you could have a lot more CPUs, and some of it could have a lot more NVIDIA GPUs,
00:36and some of it could be somebody's semi-custom ASICs.
00:39And those systems are so insanely hard to build.
00:43And they're all missing this one incredible ingredient.
00:47This incredible ingredient called NVLink.
00:50NVLink so that you could scale up these semi-custom systems and build really powerful computers.
00:56So this is the NVLink Fusion.
00:59Whether you buy completely from us, that's fantastic.
01:03Nothing gives me more joy than when you buy everything from NVIDIA.
01:08I just want you guys to know that.
01:11But it gives me tremendous joy if you just buy something from NVIDIA.
01:17And so we have some great partners.
01:25We have some great partners.
01:26L-Chip, Astero Labs, Marvell, and one of our great partners, MediaTek,
01:32are going to be partnering with us to work with ASIC or semi-custom customers, hyperscalers,
01:38who would like to build these things.
01:40Or CPU vendors who would like to build these things.
01:42And they would be their semi-custom ASIC provider.
01:46We also have Fujitsu and Qualcomm who are building their CPUs with NVLink to integrate into our ecosystem.
01:55And Cadence and Synopsys, we've worked with them to transfer our IP to them
01:59so that they can work with all of you and make that IP available to all of your chips.
02:04So this ecosystem is incredible.
02:06But this just highlights the NVLink Fusion ecosystem.
02:11Once you work with them, you instantly get integrated into the entire larger NVIDIA ecosystem
02:17that makes it possible for you to scale up into these AI supercomputers.
Be the first to comment
Add your comment

Recommended