Node Types

A wide range of GPU options are available, allowing you to select the most optimal compute resource for your workload. If GPU resources are not requested, the workload will be executed on a CPU only node.

GPU Availability

Vendor

Class

Generation

CUDA Cores

VRAM

Label

NVIDIA

Tesla V100 NVLINK

Volta

5,120

16 GB

Tesla_V100_NVLINK

NVIDIA

Tesla V100

Volta

5,120

16 GB

Tesla_V100

NVIDIA

Multi Purpose Turing

Turing

2,000+

8+ GB

NV_Turing

NVIDIA

Tesla P100

Pascal

3,584

16 GB

Tesla_P100_NVLINK

NVIDIA

Multi Purpose Pascal

Pascal

2,000+

8 GB

NV_Pascal

System Resources

Each GPU includes a certain amount of host CPU and RAM, these are included at no additional fee.

Class

vCPU

RAM

Great For

Tesla V100 NVLINK

4 Xeon Silver

32 GB

Deep learning, neural network training, HPC

Tesla V100

3

20 GB

AI inference, rendering, batch processing, hashcat

Mutli Purpose Turing

3

16 GB

Machine learning, rendering, batch processing

Tesla P100 NVLINK

4 Xeon Silver

32 GB

Entry level HPC, rendering, batch processing

Multi Purpose Pascal

1

8 GB

Video transcoding, rendering, batch processing

If a workload requests more peripheral compute resources (vCPU, RAM) than offered in a standard instance size, additional costs will incur.

Please reach out to cloud.support@coreweave.com for additional information on enhanced vCPU/RAM combinations and their costs.

CPU Availability

CPU Model

RAM per vCPU

Label

Intel Xeon v1

3 GB

xeon

AMD Epyc Rome

4 GB

epyc

Workloads without GPU requests are always scheduled on CPU nodes. If a specific CPU model is not explicitly selected, the scheduler will automatically schedule workloads requesting few CPU cores on Epyc class CPUs, as these perform exceptionally well on single thread workloads.

Requesting Compute in Kubernetes

A combination of resource requests and node affinity is used to select the type and amount of compute for your workload. CoreWeave Cloud relies only on these native Kubernetes methods for resource allocation, allowing maximum flexibilty.

Single Tesla V100
4x Tesla V100 NVLINK
Dual Pascal
16 Core Xeon CPU
Single Epyc CPU
Single Tesla V100
spec:
containers:
- name: example
resources:
limits:
cpu: 3
memory: 16Gi
nvidia.com/gpu: 1
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: gpu.nvidia.com/class
operator: In
values:
- Tesla_V100
4x Tesla V100 NVLINK
spec:
containers:
- name: example
resources:
limits:
cpu: 15
memory: 128Gi
nvidia.com/gpu: 4
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: gpu.nvidia.com/class
operator: In
values:
- Tesla_V100_NVLINK
Dual Pascal
spec:
containers:
- name: example
resources:
limits:
cpu: 2
memory: 16Gi
nvidia.com/gpu: 2
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: gpu.nvidia.com/class
operator: In
values:
- NV_Pascal
16 Core Xeon CPU
spec:
containers:
- name: example
resources:
limits:
cpu: 16
memory: 48Gi
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: cpu.coreweave.cloud/family
operator: In
values:
- xeon
Single Epyc CPU
spec:
containers:
- name: example
resources:
limits:
cpu: 1
memory: 4Gi
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: cpu.coreweave.cloud/family
operator: In
values:
- epyc

Kubernetes allows resources to be scheduled with requests and limits. When only limits are specified, the requests are set to the same amount as the limit.