verifier.org

CoreWeave

free egress, free iops, free internal transfer — behind a headline number that is eight times what most readers think it is.

$6.16/gpu/hr#large-scale-training-whe#data-movement-is-constan

the most generous data terms in the category attached to the most misread price — read it as a node rate, divide by eight, and the spot tier is genuinely competitive.

everything about data movement is free here: egress, iops, and transfer within coreweave. object storage runs $0.06 per gb monthly hot, $0.03 warm and $0.015 cold — the cheapest cold tier we found. for training pipelines that shuttle checkpoints and datasets constantly, that structure can matter more than the hourly rate.

the hourly rate needs arithmetic. coreweave prices per eight-gpu hgx node: $49.24 on-demand, which is $6.16 per gpu and expensive, and $19.71 spot, which is $2.46 per gpu and competitive. we have seen the node figure quoted as though it were per-gpu, and the error runs in the direction that makes coreweave look eight times worse than it is.

the description above is ours, condensed from the ranking. pricing moves — check it on the vendor's own page before you rely on it.

pricing
$6.16/gpu/hr
our verdict

the most generous data terms in the category attached to the most misread price — read it as a node rate, divide by eight, and the spot tier is genuinely competitive.

we researched this category against vendors' own pricing pages and licence files. that is where this line comes from — not from the vendor, and not from anything they paid for.

more gpu cloud providers

TensorDock

marketplace gpu rental with per-minute billing

#marketplace#per-minute

FluidStack

large gpu clusters brokered for training runs

#clusters#training

Shadeform

one api and console across several gpu clouds, to chase capacity

#multi-cloud#api

RunPod

we checked this$2.69/gpu/hr

cheapest verified h100 of the ten, billed by the second, with free egress and one storage trap worth knowing.

#most-workloads-it-is-the#the-billing-granularity-