Deploy

Spin up GPU compute and deploy a model in minutes.

P
Parthexpand_more
1

Choose Compute

2

Select Model

3

Configure

4

Deploy

Choose Compute

Pick the GPU type, cluster size, and region.

116

Live Summary

GPUA100 SXM × 4
RegionUS-EAST
ModelLlama 3 70B Instruct
Min replicas1
Max replicas5
Scale to zeroYes

Estimated cost

$6.60/hour

~$4752 / month at 100% utilization