The verdict
Lambda takes it, 4–1 on rounds.
On the published figures and the independent review sites, Lambda takes 4 of 6 rounds. Treat that as a starting point and check the current figures on both pages, because prices and plans move.
Choose Lambda if…
- • You want Lambda's edge on the cheapest h100 and the cheapest a100
Choose CoreWeave if…
- • You want CoreWeave's edge on how many places it runs
Round by round
- Round 1The cheapest H100Lambda winsCoreWeave5/10
$6.16
per GPU, worked out from $49.24 an hour for an HGX H100 with eight GPUs
- $6.16, against $3.99
Source: coreweave.com ↗
Lambda wins this round on what the company itself publishes.
- Round 2The cheapest A100Lambda winsCoreWeave5/10
$2.70
per GPU, worked out from $21.60 an hour for an eight-A100 instance
- $2.70, against $1.99
Source: coreweave.com ↗
Lambda wins this round on what the company itself publishes.
- Round 3The cheapest B200Lambda winsCoreWeave5/10
$8.60
per GPU, worked out from $68.80 an hour for an HGX B200 with eight GPUs
- $8.60, against $6.69
Source: coreweave.com ↗
Lambda wins this round on what the company itself publishes.
- Round 4How cheap you can startLambda winsLambda8/10
$0.79
NVIDIA Tesla V100
- the cheapest GPU on its price list, a Quadro RTX 6000
Source: lambda.ai ↗
CoreWeave5/10$1.25
the L40, worked out from $10.00 an hour for eight
- $1.25, against $0.69
Source: coreweave.com ↗
Lambda wins this round on what the company itself publishes.
- Round 5How many places it runsCoreWeave winsLambda5/10
14 regions
on-demand cloud regions, though not every instance type is in every region
- 14 regions, against 51 data centers
Source: docs.lambda.ai ↗
CoreWeave8/1051 data centers
operated across North America and Europe, counted as data centers rather than regions
Source: coreweave.com ↗
CoreWeave wins this round on what the company itself publishes.
- Round 6What it costs to get your data outTieCoreWeave7/10
Free
data transfer between CoreWeave and the internet is listed as free
Source: coreweave.com ↗
Nothing to separate them here.
Side by side
- the cheapest GPU on its price list, a Quadro RTX 6000
- A100, PCIe and SXM
- per GPU, worked out from $21.60 an hour for an eight-A100 instance
- on-demand GPU instances, North America
- $3.99, against $1.99
- $1.99, against $1.19
- $6.69, against $5.98
- $0.69, against $0.34
- $6.16, against $1.99
- $2.70, against $1.19
- $6.31, against $3.59
- $8.60, against $5.98