GPU compute

GPU offers, priced by the hour.

Compare GPU memory, server resources and rental costs. Narrow the list to the hardware you need, then check storage and billing before choosing an offer.

Offers: 91
Providers: 9
GPU models / variants: 51
Largest GPU memory: 288 GB

13 offers match your selection. Prices use the currency shown; currencies are not converted.

GPUGPU memorySystemRegionPer hour

OVHcloud · On Demand

2×NVIDIA L4

OVHcloud US l4-180

48 GB 24 GB each

180 GB45 vCPU

Not reported

$2 $1 per GPU $0.042 per GB VRAM

View official offer

Storage and billing details

Included storage: 400 GB NVMe
Billing: Hourly instance rate in US dollars, before taxes. The displayed rate covers the complete listed configuration, not each GPU separately. Monthly plans are a different purchase choice.
Storage costs and persistence: Extra volumes, backups and other services are separate budget items. Keep important data outside the running instance’s local disks.
Configuration: Two 24 GB L4 cards with additional host resources. Confirm multi-GPU runtime support before treating the pair as suitable for one large model. Select a supported US location and check capacity in the order flow.

OVHcloud · On Demand

NVIDIA L4

OVHcloud US l4-90

24 GB

90 GB22 vCPU

Not reported

$1 per hour $0.042 per GB VRAM

View official offer

Storage and billing details

Included storage: 400 GB NVMe
Billing: Hourly instance rate in US dollars, before taxes. The displayed rate covers the complete listed configuration, not each GPU separately. Monthly plans are a different purchase choice.
Storage costs and persistence: Extra volumes, backups and other services are separate budget items. Keep important data outside the running instance’s local disks.
Configuration: A single 24 GB L4 with 90 GB of host RAM. GPU memory and system memory serve different parts of the workload; a large host does not make the GPU hold a larger model. Select a supported US location and check capacity in the order flow.

Runpod · On Demand

NVIDIA L4

Runpod Pod · NVIDIA L4

24 GB per GPU · count unconfirmed

50 GB12 vCPU

Not reported

$0.59 per hour

View official offer

Storage and billing details

Billing: Pod compute is billed per second; choose the exact configuration in the provider console.
Storage costs and persistence: Storage is separate. Volume disk costs $0.10/GB-month while running and $0.20 while stopped. Network volume rates depend on size and storage tier.
Configuration: Choose the cloud and number of GPUs in the deployment console. The rate shown does not guarantee available capacity.

Scaleway · On Demand

NVIDIA L4

Scaleway L4-1-24G

24 GB

48 GB8 vCPU

Not reported

€0.79 per hour €0.033 per GB VRAM

View official offer

Scaleway · On Demand

8×NVIDIA L4

Scaleway L4-8-24G

192 GB 24 GB each

384 GB64 vCPU

Not reported

€6.3 €0.788 per GPU €0.033 per GB VRAM

View official offer

Scaleway · On Demand

4×NVIDIA L4

Scaleway L4-4-24G

96 GB 24 GB each

192 GB32 vCPU

Not reported

€3.15 €0.788 per GPU €0.033 per GB VRAM

View official offer

Scaleway · On Demand

2×NVIDIA L4

Scaleway L4-2-24G

48 GB 24 GB each

96 GB16 vCPU

Not reported

€1.58 €0.79 per GPU €0.033 per GB VRAM

View official offer

Runpod · On Demand

NVIDIA L40

Runpod Pod · NVIDIA L40

48 GB per GPU · count unconfirmed

94 GB8 vCPU

Not reported

$0.82 per hour

View official offer

Storage and billing details

Billing: Pod compute is billed per second at the hourly rate shown. Check the final configuration price before deploying.
Storage costs and persistence: Storage is billed separately. Persistent volume disks continue accruing charges while a Pod is stopped.
Configuration: Choose the cloud, region and number of GPUs in the deployment console. Available configurations and stock can vary.

DigitalOcean · On Demand

NVIDIA L40S

L40S GPU Droplet · 1 GPU

48 GB

64 GiB8 vCPU

Not reported

$1.57 per hour $0.033 per GB VRAM

View official offer

Storage and billing details

Included storage: 500 GiB NVMe boot disk
Transfer allowance: 10,000 GiB transfer allowance; excess outbound transfer is $0.01/GiB, pooled at team level.
Billing: Per-second compute billing. Charges continue while powered off until the Droplet is destroyed. Confirm the minimum charge before ordering.
Storage costs and persistence: Keep recovery copies outside the instance.

OVHcloud · On Demand

2×NVIDIA L40S

OVHcloud US l40s-180

96 GB 48 GB each

180 GB30 vCPU

Not reported

$3.6 $1.8 per GPU $0.038 per GB VRAM

View official offer

Storage and billing details

Included storage: 400 GB NVMe
Billing: Hourly instance rate in US dollars, before taxes. The displayed rate covers the complete listed configuration, not each GPU separately. Monthly plans are a different purchase choice.
Storage costs and persistence: Extra volumes, backups and other services are separate budget items. Keep important data outside the running instance’s local disks.
Configuration: Two L40S cards provide 48 GB each. More devices can serve separate jobs or a supported distributed workload; they do not automatically form a single memory pool. Select a supported US location and check capacity in the order flow.

OVHcloud · On Demand

NVIDIA L40S

OVHcloud US l40s-90

48 GB

90 GB15 vCPU

Not reported

$1.8 per hour $0.038 per GB VRAM

View official offer

Storage and billing details

Included storage: 400 GB NVMe
Billing: Hourly instance rate in US dollars, before taxes. The displayed rate covers the complete listed configuration, not each GPU separately. Monthly plans are a different purchase choice.
Storage costs and persistence: Extra volumes, backups and other services are separate budget items. Keep important data outside the running instance’s local disks.
Configuration: A single 48 GB L40S for supported inference, graphics or rendering software. Leave memory for runtime work and context alongside model weights. Select a supported US location and check capacity in the order flow.

Runpod · On Demand

NVIDIA L40S

Runpod Pod · NVIDIA L40S

48 GB per GPU · count unconfirmed

94 GB16 vCPU

Not reported

$1.09 per hour

View official offer

Storage and billing details

Billing: Pod compute is billed per second. Check the selected cloud and configuration rate before deploying.
Storage costs and persistence: Storage is extra. Container disks are $0.10/GB/month; volume disks are $0.10/GB/month while running and $0.20/GB/month while stopped.
Configuration: The rate is per GPU-hour. Choose the number of GPUs and region in the console, then check capacity and the total before deploying.

Vultr · On Demand

NVIDIA L40S

Vultr NVIDIA L40S · 1 GPU

48 GB

180 GB16 vCPU

Not reported

$1.671 per hour $0.035 per GB VRAM

View official offer

VRAM is one input, not a guarantee.

Driver versions, model architecture, quantization and context length all change what fits. Several GPUs do not automatically behave like one larger GPU; your runtime must support splitting the workload. Check the model and software requirements alongside memory.

Check a model's memory