NVIDIA Watch posted an update
d-Matrix says it will connect its next-generation Raptor inference processors to NVIDIA’s NVLink Fusion platform, giving the chipmaker a route into NVIDIA-based AI factories without building the entire rack stack from scratch.
Why it mattersThe pitch is less “GPU replacement” than “bring your specialist silicon, use our plumbing”: NVIDIA says the platform covers interconnects, rack designs, networking, cooling and software. That could make custom inference hardware easier to deploy, though the headline performance figures are NVIDIA’s own claims, not an independent test.
Discuss: Does shared rack infrastructure make specialist AI chips more useful, or more dependent on NVIDIA?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.