3 ms·
The RTX 4090 has 330.3 TOPS with INT8 precision, so for inference workloads it is still a magnitude faster
by mosshammer 3y ago
The RTX 4090 has 330.3 TOPS with INT8 precision, so for inference workloads it is still a magnitude faster
loading thread…