€432/mo
- 2× AMD EPYC 7413 · 48 cores / 96 threads
- 128 GB RAM
- 2x960GB SATA
- 1× L40S
- 1 Gbps · 30 TB traffic
GPU
L40S servers
Where to rent a server with the L40S, and for how much. Prices excl. VAT.
- Cheapest
- €432
- per month
- Servers
- 6
- Providers
- 1
Cheapest L40S servers
In-stock servers first, by monthly price incl. setup fee.
- €482/mo
- 2× AMD EPYC 7543 · 64 cores / 128 threads
- 128 GB RAM
- 2x960GB SATA
- 1× L40S
- 1 Gbps · 30 TB traffic
- €555/mo
- 2× AMD EPYC 9224 · 48 cores / 96 threads
- 128 GB RAM
- 2x960GB SATA
- 1× L40S
- 1 Gbps · 30 TB traffic
- €558/mo
- 2× Intel Xeon 4314 · 32 cores / 64 threads
- 128 GB RAM
- 2x960GB NVMe
- 1× L40S
- 1 Gbps · 30 TB traffic
- €605/mo
- 2× AMD EPYC 9334 · 64 cores / 128 threads
- 128 GB RAM
- 2x960GB SATA
- 1× L40S
- 1 Gbps · 30 TB traffic
- €857/mo
- 2× AMD EPYC 9334 · 64 cores / 128 threads
- 128 GB RAM
- 2x960GB SATA
- 2× L40S
- 1 Gbps · 30 TB traffic
Providers offering the L40S
Benchmarks
| Suite | Metric | Result | Source |
|---|---|---|---|
| blender | blender_gpu_spm_3scenes | 7,848.06 | blender · retrieved |
| mlperf_inference | v5.0:3d-unet-99.9:offline:per_accelerator_derived | 3.45 | mlperf · retrieved |
| mlperf_inference | v5.0:3d-unet-99:offline:per_accelerator_derived | 3.67 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99.9:offline:per_accelerator_derived | 12,564.6 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99.9:server:per_accelerator_derived | 11,873.7 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99:offline:per_accelerator_derived | 28,916.3 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99:server:per_accelerator_derived | 24,795.8 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99.9:offline:per_accelerator_derived | 860.66 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99.9:server:per_accelerator_derived | 814.65 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99:offline:per_accelerator_derived | 868.34 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99:server:per_accelerator_derived | 852.64 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99.9:offline:per_accelerator_derived | 446.09 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99.9:server:per_accelerator_derived | 398.92 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99:offline:per_accelerator_derived | 446.09 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99:server:per_accelerator_derived | 398.92 | mlperf · retrieved |
| mlperf_inference | v5.0:resnet:offline:per_accelerator_derived | 42,237.7 | mlperf · retrieved |
| mlperf_inference | v5.0:resnet:server:per_accelerator_derived | 42,756.2 | mlperf · retrieved |
| mlperf_inference | v5.0:retinanet:offline:per_accelerator_derived | 801.71 | mlperf · retrieved |
| mlperf_inference | v5.0:retinanet:server:per_accelerator_derived | 761.93 | mlperf · retrieved |
| mlperf_inference | v5.0:stable-diffusion-xl:offline:per_accelerator_derived | 0.77 | mlperf · retrieved |
| mlperf_inference | v5.0:stable-diffusion-xl:server:per_accelerator_derived | 0.63 | mlperf · retrieved |
| mlperf_inference | v5.1:llama3.1-8b:offline:per_accelerator_derived | 1,642.22 | mlperf · retrieved |
| mlperf_inference | v5.1:llama3.1-8b:server:per_accelerator_derived | 1,207.14 | mlperf · retrieved |
| mlperf_inference | v5.1:retinanet:offline:per_accelerator_derived | 814.25 | mlperf · retrieved |
| mlperf_inference | v5.1:retinanet:server:per_accelerator_derived | 762.67 | mlperf · retrieved |
| mlperf_inference | v5.1:stable-diffusion-xl:offline:per_accelerator_derived | 0.69 | mlperf · retrieved |
| mlperf_inference | v5.1:stable-diffusion-xl:server:per_accelerator_derived | 0.74 | mlperf · retrieved |
| mlperf_inference | v5.1:whisper:offline:per_accelerator_derived | 1,159 | mlperf · retrieved |
| mlperf_inference | v6.0:whisper:offline:per_accelerator_derived | 1,823.45 | mlperf · retrieved |
| blender | blender_gpu_spm_3scenes | 7,848.06 | blender · retrieved |
| mlperf_inference | v5.0:3d-unet-99.9:offline:per_accelerator_derived | 3.45 | mlperf · retrieved |
| mlperf_inference | v5.0:3d-unet-99:offline:per_accelerator_derived | 3.67 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99.9:offline:per_accelerator_derived | 12,564.6 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99.9:server:per_accelerator_derived | 11,873.7 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99:offline:per_accelerator_derived | 28,916.3 | mlperf · retrieved |
| mlperf_inference | v5.0:dlrm-v2-99:server:per_accelerator_derived | 24,795.8 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99.9:offline:per_accelerator_derived | 860.66 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99.9:server:per_accelerator_derived | 814.65 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99:offline:per_accelerator_derived | 868.34 | mlperf · retrieved |
| mlperf_inference | v5.0:gptj-99:server:per_accelerator_derived | 852.64 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99.9:offline:per_accelerator_derived | 446.09 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99.9:server:per_accelerator_derived | 398.92 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99:offline:per_accelerator_derived | 446.09 | mlperf · retrieved |
| mlperf_inference | v5.0:llama2-70b-99:server:per_accelerator_derived | 398.92 | mlperf · retrieved |
| mlperf_inference | v5.0:resnet:offline:per_accelerator_derived | 42,237.7 | mlperf · retrieved |
| mlperf_inference | v5.0:resnet:server:per_accelerator_derived | 42,756.2 | mlperf · retrieved |
| mlperf_inference | v5.0:retinanet:offline:per_accelerator_derived | 801.71 | mlperf · retrieved |
| mlperf_inference | v5.0:retinanet:server:per_accelerator_derived | 761.93 | mlperf · retrieved |
| mlperf_inference | v5.0:stable-diffusion-xl:offline:per_accelerator_derived | 0.77 | mlperf · retrieved |
| mlperf_inference | v5.0:stable-diffusion-xl:server:per_accelerator_derived | 0.63 | mlperf · retrieved |
| mlperf_inference | v5.1:llama3.1-8b:offline:per_accelerator_derived | 1,642.22 | mlperf · retrieved |
| mlperf_inference | v5.1:llama3.1-8b:server:per_accelerator_derived | 1,207.14 | mlperf · retrieved |
| mlperf_inference | v5.1:retinanet:offline:per_accelerator_derived | 814.25 | mlperf · retrieved |
| mlperf_inference | v5.1:retinanet:server:per_accelerator_derived | 762.67 | mlperf · retrieved |
| mlperf_inference | v5.1:stable-diffusion-xl:offline:per_accelerator_derived | 0.69 | mlperf · retrieved |
| mlperf_inference | v5.1:stable-diffusion-xl:server:per_accelerator_derived | 0.74 | mlperf · retrieved |
| mlperf_inference | v5.1:whisper:offline:per_accelerator_derived | 1,159 | mlperf · retrieved |
| mlperf_inference | v6.0:whisper:offline:per_accelerator_derived | 1,823.45 | mlperf · retrieved |