7 ms·

Have you tried running really large models with this? And how are you managing allocating resources ?