# CoreWeave vs Lambda

Canonical: https://inetgeek.com/compare/coreweave-vs-lambda/

Every value below is read from CoreWeave's and Lambda's own documentation. See https://inetgeek.com/methodology/ for how.

## At a glance

| Criterion | CoreWeave | Lambda |
| --- | --- | --- |
| Egress over allowance | Free. CoreWeave charges nothing for egress or for I/O operations, and runs a zero-egress migration programme for moving data in. | Free. Lambda states transparent pricing with no egress fees. |
| H100 SXM, per GPU-hour | $6.16 per GPU-hour for H100 SXM 80GB. CoreWeave sells the 8-GPU HGX H100 node at $49.24/hour on demand — the per-GPU figure is that divided by eight, ours rather than CoreWeave's, and you cannot rent one card. | $4.29 per GPU-hour for H100 SXM 80GB on a single-GPU instance, falling to $3.99 on an 8-GPU node. The PCIe variant is $3.29. |
| Cheapest GPU, per hour | $1.25 per GPU-hour for L40, again as an 8-GPU node at $10.00/hour. A100 80GB nodes are $21.60/hour, or $2.70 per GPU. | $0.69 per GPU-hour for a Quadro RTX 6000 with 24GB; the V100 is $0.79 and the A6000 $1.09. |
| Largest GPU offered | 288GB per GPU on B300-class hardware; the largest listed on-demand node is HGX B200 at 180GB per GPU, eight GPUs, $68.80/hour. | 180GB per GPU on B200 SXM6, from $6.69 per GPU-hour on an 8-GPU node to $6.99 on a single GPU. |

## Where they differ

### Egress over allowance

- CoreWeave: Free. CoreWeave charges nothing for egress or for I/O operations, and runs a zero-egress migration programme for moving data in. ([source](https://www.coreweave.com/pricing))
- Lambda: Free. Lambda states transparent pricing with no egress fees. ([source](https://lambda.ai/service/gpu-cloud))

### H100 SXM, per GPU-hour

- CoreWeave: $6.16 per GPU-hour for H100 SXM 80GB. CoreWeave sells the 8-GPU HGX H100 node at $49.24/hour on demand — the per-GPU figure is that divided by eight, ours rather than CoreWeave's, and you cannot rent one card. ([source](https://www.coreweave.com/pricing))
- Lambda: $4.29 per GPU-hour for H100 SXM 80GB on a single-GPU instance, falling to $3.99 on an 8-GPU node. The PCIe variant is $3.29. ([source](https://lambda.ai/service/gpu-cloud))

### Cheapest GPU, per hour

- CoreWeave: $1.25 per GPU-hour for L40, again as an 8-GPU node at $10.00/hour. A100 80GB nodes are $21.60/hour, or $2.70 per GPU. ([source](https://www.coreweave.com/pricing))
- Lambda: $0.69 per GPU-hour for a Quadro RTX 6000 with 24GB; the V100 is $0.79 and the A6000 $1.09. ([source](https://lambda.ai/service/gpu-cloud))

### Largest GPU offered

- CoreWeave: 288GB per GPU on B300-class hardware; the largest listed on-demand node is HGX B200 at 180GB per GPU, eight GPUs, $68.80/hour. ([source](https://www.coreweave.com/pricing))
- Lambda: 180GB per GPU on B200 SXM6, from $6.69 per GPU-hour on an 8-GPU node to $6.99 on a single GPU. ([source](https://lambda.ai/service/gpu-cloud))

## Which should you choose?

Pick CoreWeave if Training runs that want a whole HGX node and move large datasets — free egress and free I/O operations remove the charge that usually dominates a data-heavy job.

Pick Lambda if Renting exactly the instance shape you need — Lambda publishes 1x, 2x, 4x and 8x plans for the same card, so a single H100 is a real product rather than a fraction of a node.

Consider something else: CoreWeave — You cannot rent a single card: the H100 is sold as an eight-GPU node at $49.24 an hour, so a one-GPU experiment costs eight GPUs.

Consider something else: Lambda — Its H100 SXM is the most expensive per GPU-hour in this set at $4.29 single-card, and there is no published reserved discount to bring it down.

## Documented by only one

- Storage price: CoreWeave: $0.06 per GB per month for hot AI object storage, and $0.070 per GB per month for the tier listed beside it.
- Billing granularity: Lambda: By the minute.
- Reserved pricing: CoreWeave: Up to 60% below on-demand for reserved compute capacity.

## Questions this comparison answers

**Egress over allowance: CoreWeave or Lambda?**

CoreWeave: Free. CoreWeave charges nothing for egress or for I/O operations, and runs a zero-egress migration programme for moving data in.
Lambda: Free. Lambda states transparent pricing with no egress fees.

**Should I choose CoreWeave or Lambda?**

Pick CoreWeave if Training runs that want a whole HGX node and move large datasets — free egress and free I/O operations remove the charge that usually dominates a data-heavy job.
Pick Lambda if Renting exactly the instance shape you need — Lambda publishes 1x, 2x, 4x and 8x plans for the same card, so a single H100 is a real product rather than a fraction of a node.