H100 SXM vs L40S: Which GPU Should You Choose?

H100 SXM offers more device memory and NVLink than L40S; check whether either difference affects your workload before comparing rates.

Updated 2026-10-07 · refreshed every 12 hours

Choose H100 SXM when its larger memory or a verified NVLink configuration is required. For a workload that fits L40S without those links, compare both GPUs using the current rates and measured completion cost.

Run a bounded test with the same model, precision, batch size and sequence length on each feasible option. Record peak memory, billed runtime and completed work; use the actual checkout rate to compare compute cost per completed job.

Does the job exceed L40S memory or require NVLink? If yes, verify that an H100 SXM rental meets both the memory and topology requirements. If no, benchmark both fitting devices at current checkout rates.

Which option meets your deadline and job budget in a representative test? Measure completion time first. The H100 SXM-to-L40S modeled cost-per-job ratio is 1.46; it uses specification-based assumptions and does not establish real runtime or a deadline.

Can the exact configuration launch under the rental terms your job needs? Confirm the GPU count, topology and availability at checkout. Use spot only after testing checkpoint recovery; if the required configuration cannot launch, reassess fit before substituting another GPU.

Today's prices

USD/hr · 1x GPU · prices grouped by provider and purchase model

Provider	H100 SXM	L40S
Hyperstack	$3.20 · Not reported	Not offered
RunPod	$3.49	$1.09
DataCrunch	$3.85	$1.59
Vast.ai	$3.87	$0.80
Crusoe	$3.90 · Not reported	$1.50 · Not reported
Together AI	$3.99 · Not reported	Not offered
Paperspace	$5.95 · Not reported	Not offered
Lambda	Unavailable as of 2026-10-07	Not offered

Configuration prices used by workload rows

USD/hr · currently eligible on-demand configurations at the exact GPU count shown in the workload table

Provider	GPU	Configuration	Price	Price basis	Availability
Lambda	H100 SXM	2 GPUs	$8.38/hr	Published list rate	In stock

see all H100 prices →

Where the two parts differ

Specification	H100 SXM	L40S
Memory	80 GB HBM3	48 GB GDDR6
Memory bandwidth	3350 GB/s	864 GB/s
Dense BF16	989 TFLOPS	362 TFLOPS
Dense FP8	1979 TFLOPS	733 TFLOPS
NVLink	900 GB/s	No NVLink
TDP	700 W	350 W

Prices are refreshed every 12 hours. Marlin matches your requirements across supported providers. Try Marlin →

4.00×

H100 SXM-to-L40S on-demand hourly price ratio. Below one favors the first GPU on hourly rate; above one favors the second.

1.46×

H100 SXM-to-L40S modeled cost-per-job ratio. Below one favors the first under the stated model; validate with measured runtime.

$2336/mo

Continuous-use monthly baseline for one H100 SXM at its current on-demand minimum; scale the budget to your billed hours.

What your workload needs

Find your parameter count on the model's Hugging Face card. Serving maps to the inference rows, training to the fine-tuning rows.

Workload	VRAM needed	H100 SXM	L40S	Cheapest today
7B Q4 inference	4 GB	1-GPU configuration: $3.20/hr (Not reported; observed 2026-10-07; price basis: Published list rate)	1-GPU configuration: $0.80/hr (In stock; observed 2026-10-07; price basis: Marketplace quote)	L40S: 1 GPU at $0.8/hr
13B Q4 inference	8 GB	1-GPU configuration: $3.20/hr (Not reported; observed 2026-10-07; price basis: Published list rate)	1-GPU configuration: $0.80/hr (In stock; observed 2026-10-07; price basis: Marketplace quote)	L40S: 1 GPU at $0.8/hr
70B QLoRA	46 GB	1-GPU configuration: $3.20/hr (Not reported; observed 2026-10-07; price basis: Published list rate)	1-GPU configuration: $0.80/hr (In stock; observed 2026-10-07; price basis: Marketplace quote)	L40S: 1 GPU at $0.8/hr
7B full fine-tune	112 GB	2-GPU configuration: $8.38/hr (In stock; observed 2026-10-07; price basis: Published list rate)	No eligible 3-GPU configuration	H100 SXM: 2 GPUs at $8.38/hr
70B FP8 training	154 GB	2-GPU configuration: $8.38/hr (In stock; observed 2026-10-07; price basis: Published list rate)	No eligible 4-GPU configuration	H100 SXM: 2 GPUs at $8.38/hr

What the numbers say

H100 SXM has a larger memory budget and NVLink support; L40S does not provide NVLink. These are configuration constraints before they are price arguments. If the workload fits one L40S, a multi-GPU interconnect advantage may be irrelevant; if it spans GPUs, verify how the offered machine connects them.

H100 SXM has an on-demand minimum of $3.20/hr and L40S has an on-demand minimum of $0.80/hr. The H100 SXM-to-L40S on-demand price ratio is 4. When both configurations meet the same requirements, compare the current rates rather than assuming that either GPU is always cheaper. Spot listings have different interruption terms and are not the basis of that ratio.

Measure a representative serving or training run on each feasible option. Record peak memory, billed time and completed work, including recovery overhead if you evaluate spot capacity. Do not move an NVLink-dependent job to L40S merely because a listing is available.

How to test the comparison

Hold the workload constant
Compare the same model, precision, batch size, sequence length and output-quality target. Changing these between GPUs changes the question being tested.
Measure the result you need
For serving, record latency and throughput at the intended concurrency. For training, record time for the same completed work. Check peak memory and failures in both cases.
Price the complete run
Multiply the full configuration's checkout rate by billed runtime. Compare storage, transfers and restart costs separately. A datasheet-based ratio is a screening model, not this measurement.

Before you rent

Data egress can raise the total bill beyond GPU hours; verify the provider's network pricing for your region and expected transfer volume.

Exceeding single-GPU VRAM can force a multi-GPU configuration, increasing GPU count and communication overhead.

Spot rows marked interruptible expose jobs to restarts, which can erase rate savings when the workload cannot resume from a checkpoint.

Billing granularity can make a short experiment cost more than its runtime suggests; find the minimum charge on each provider's pricing page when estimating the billed amount.

Marlin matches your workload requirements to the lowest-priced suitable GPU across supported providers.

Memory fit and cost-per-job comparisons are models. Datasheet peaks do not establish application throughput; verify the listed configuration with your own workload. https://www.tensordock.com/host-pricing · https://www.coreweave.com/pricing · https://vast.ai/pricing · https://datacrunch.io/pricing · https://resources.nvidia.com/en-us-tensor-core/nvidia-tensor-core-gpu-datasheet · https://resources.nvidia.com/en-us-l40s/l40s-datasheet-28413 · https://www.runpod.io/pricing · https://lambda.ai/service/gpu-cloud · https://www.hyperstack.cloud/gpu-pricing · https://www.together.ai/pricing · https://crusoe.ai/cloud/pricing

Marlin beta

Stop comparing. Start running.

Marlin matches your workload to the lowest-priced suitable GPU across supported providers.

Get started with Marlin