atlas
vt1…
Try

3 of 1173 instances · cheapest is vt1.3xlarge at $0.7341/hr

Instance Workload vCPUs Memory Local disk Network On-demand/hr ↑ Per vCPU
vt1.3xlargeGPU & accelerators 12 24 GiB EBS only 3.125 Gbps $0.7341 $0.0612
vt1.6xlargeGPU & accelerators 24 48 GiB EBS only 6.25 Gbps $1.468 $0.0612
vt1.24xlargeGPU & accelerators 96 192 GiB EBS only 25 Gbps $5.873 $0.0612