3 ms·
Nvidia have less moat for inference workloads since inference is modular. AMD would be mistaken to go after training workloads but that's not what they're going
by energy123 2y ago
Nvidia have less moat for inference workloads since inference is modular. AMD would be mistaken to go after training workloads but that's not what they're going after.