MarketplaceFind GPUs
HelpCreate account
GPU MARKETPLACE 47 models

Find the right GPU.

Start with your workload. Compare the resources that matter.

Help me choose

What are you building?

Quick starts
LLM inference24 GB+ · vLLMImage generation16 GB+ · ComfyUIModel training48 GB+ · PyTorchStart for lessLowest hourly cost
GPUs per node
Pricing plan
PyTorch 2.5 · CUDA 12.40 GB diskCustomize
Explore templates

47 configurations

USD · entire node / hour
V100 16GB1×
Volta
VRAM / GPU16GBHBM2
$0.080/hr$1.92 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesVirginia+116 available nodesDetails
Memory bandwidth
900 GB/s
Interconnect
PCIe Gen3, 32 GB/s
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0013 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesVirginiaUS EastGermanyFrankfurtEU Central
VRAM / GPU96GBGDDR7
$0.390/hr$9.36 / day
Configure
16 vCPU128 GB RAM1 TB NVMeUnited StatesOregon+124 available nodesDetails
Memory bandwidth
1,597 GB/s
Interconnect
PCIe Gen5 x16
Local storage
1,000 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0065 / minute
GPU memory
96 GB per GPU · 96 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
VRAM / GPU48GBGDDR7
$0.290/hr$6.96 / day
Configure
12 vCPU96 GB RAM1 TB NVMeUnited StatesOregon+116 available nodesDetails
Memory bandwidth
1,344 GB/s
Interconnect
PCIe 5.0 x16
Local storage
1,000 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0048 / minute
GPU memory
48 GB per GPU · 48 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
VRAM / GPU32GBGDDR7
$0.190/hr$4.56 / day
Configure
10 vCPU64 GB RAM500 GB NVMeUnited StatesOregon16 available nodesDetails
Memory bandwidth
896 GB/s
Interconnect
PCIe 5.0 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0032 / minute
GPU memory
32 GB per GPU · 32 GB total
Configure in a regionUnited StatesOregonUS West
VRAM / GPU24GBGDDR7
$0.150/hr$3.60 / day
Configure
8 vCPU48 GB RAM500 GB NVMeUnited StatesOregon16 available nodesDetails
Memory bandwidth
672 GB/s
Interconnect
PCIe 5.0 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0025 / minute
GPU memory
24 GB per GPU · 24 GB total
Configure in a regionUnited StatesOregonUS West
RTX 6000 Ada1×
Ada Lovelace
VRAM / GPU48GBGDDR6
$0.190/hr$4.56 / day
Configure
12 vCPU96 GB RAM1 TB NVMeUnited StatesOregon+124 available nodesDetails
Memory bandwidth
960 GB/s
Interconnect
PCIe 4.0 x16; NVLink: No
Local storage
1,000 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0032 / minute
GPU memory
48 GB per GPU · 48 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 5000 Ada1×
Ada Lovelace
VRAM / GPU32GBGDDR6
$0.170/hr$4.08 / day
Configure
10 vCPU64 GB RAM500 GB NVMeUnited StatesOregon+116 available nodesDetails
Memory bandwidth
Not specified
Interconnect
PCIe 4.0 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0028 / minute
GPU memory
32 GB per GPU · 32 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 4500 Ada1×
Ada Lovelace
VRAM / GPU24GBGDDR6
$0.150/hr$3.60 / day
Configure
8 vCPU48 GB RAM500 GB NVMeUnited StatesOregon16 available nodesDetails
Memory bandwidth
432 GB/s
Interconnect
PCIe 4.0 x16; NVLink: No
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0025 / minute
GPU memory
24 GB per GPU · 24 GB total
Configure in a regionUnited StatesOregonUS West
RTX 4000 Ada1×
Ada Lovelace
VRAM / GPU20GBGDDR6
$0.090/hr$2.16 / day
Configure
8 vCPU48 GB RAM500 GB NVMeUnited StatesOregon+116 available nodesDetails
Memory bandwidth
360 GB/s
Interconnect
PCIe 4.0 x16; NVLink: No
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
20 GB per GPU · 20 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX A60001×
AmpereNVLink
VRAM / GPU48GBGDDR6
$0.170/hr$4.08 / day
Configure
12 vCPU96 GB RAM1 TB NVMeUnited StatesOregon+124 available nodesDetails
Memory bandwidth
768 GB/s
Interconnect
PCIe Gen4 x16; NVLink 2-way (2-slot/3-slot bridges)
Local storage
1,000 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0028 / minute
GPU memory
48 GB per GPU · 48 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX A50001×
AmpereNVLink
VRAM / GPU24GBGDDR6
$0.090/hr$2.16 / day
Configure
8 vCPU48 GB RAM500 GB NVMeUnited StatesOregon+124 available nodesDetails
Memory bandwidth
768 GB/s
Interconnect
PCIe 4.0 x16; NVLink 112.5 GB/s bidirectional
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
24 GB per GPU · 24 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX A45001×
AmpereNVLink
VRAM / GPU20GBGDDR6
$0.090/hr$2.16 / day
Configure
8 vCPU48 GB RAM500 GB NVMeUnited StatesOregon16 available nodesDetails
Memory bandwidth
640 GB/s
Interconnect
PCIe 4.0 x16; NVLink 112.5 GB/s bidirectional
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
20 GB per GPU · 20 GB total
Configure in a regionUnited StatesOregonUS West
RTX A40001×
AmpereLowest price
VRAM / GPU16GBGDDR6
$0.050/hr$1.20 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesOregon+124 available nodesDetails
Memory bandwidth
448 GB/s
Interconnect
PCIe 4.0 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0008 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 50901×
Blackwell
VRAM / GPU32GBGDDR7
$0.190/hr$4.56 / day
Configure
8 vCPU64 GB RAM500 GB NVMeUnited StatesOregon+164 available nodesDetails
Memory bandwidth
1,792 GB/s
Interconnect
PCIe Gen5 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0032 / minute
GPU memory
32 GB per GPU · 32 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 50801×
Blackwell
VRAM / GPU16GBGDDR7
$0.090/hr$2.16 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesOregon+132 available nodesDetails
Memory bandwidth
960 GB/s
Interconnect
PCIe Gen5 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 5070 Ti1×
Blackwell
VRAM / GPU16GBGDDR7
$0.090/hr$2.16 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesOregon24 available nodesDetails
Memory bandwidth
896 GB/s
Interconnect
PCIe Gen5 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesOregonUS West
RTX 40901×
Ada Lovelace
VRAM / GPU24GBGDDR6X
$0.090/hr$2.16 / day
Configure
8 vCPU48 GB RAM500 GB NVMeUnited StatesOregon+164 available nodesDetails
Memory bandwidth
1,008 GB/s
Interconnect
PCIe Gen4 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
24 GB per GPU · 24 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 4080 SUPER1×
Ada Lovelace
VRAM / GPU16GBGDDR6X
$0.090/hr$2.16 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesOregon+132 available nodesDetails
Memory bandwidth
Not specified
Interconnect
PCIe Gen4 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 40801×
Ada Lovelace
VRAM / GPU16GBGDDR6X
$0.090/hr$2.16 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesOregon+132 available nodesDetails
Memory bandwidth
717 GB/s
Interconnect
PCIe Gen4 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesOregonUS WestGermanyFrankfurtEU Central
RTX 4070 Ti SUPER1×
Ada Lovelace
VRAM / GPU16GBGDDR6X
$0.090/hr$2.16 / day
Configure
6 vCPU32 GB RAM500 GB NVMeUnited StatesOregon24 available nodesDetails
Memory bandwidth
Not specified
Interconnect
PCIe Gen4 x16
Local storage
500 GB included
Network
100 Gbps
Persistent disk
None selected
Software
PyTorch 2.5 · CUDA 12.4
Billing
$0.0015 / minute
GPU memory
16 GB per GPU · 16 GB total
Configure in a regionUnited StatesOregonUS West
21–40 of 47 configurations

Browse freely. Sign in when you are ready to order.

SIDE BY SIDE

Compare configurations

Prices are saved when selected. Review the current quote before confirming an order.

Confirm action