atlas
p5en…
Try

1 of 1348 instances · cheapest is p5en.48xlarge at $63.30/hr

Instance Workload vCPUs Memory Local disk Network On-demand/hr ↑ Per vCPU
p5en.48xlargeGPU & accelerators 192 2 TiB 31 TB NVMe 3200 Gbps $63.30 $0.3297