Runpod Our sponsor
USA 4 configs 1x-2x
On-Demand
from $0.19
Nvidia V100
Legacy data center GPU with FP64 support for scientific computing.
Last update 6 minutes ago
Aggregating historical prices...
Key Specifications
V100 Pricing and Availability
Barriers to entry are low here. 21 cloud providers list the V100 with prices starting at just $0.09/hr per GPU (on-demand), making it accessible for testing without a large commitment. Spot instances start lower, at $0.05/hr per GPU.
Vast.ai In stock
USA 13 configs 1x-4x
On-Demand
from $0.09
Reserved
from $0.16
Spot
from $0.05
Theta EdgeCloud In stock
USA 2 configs 1x-2x
On-Demand
from $0.99
Runcrate In stock
USA 2 configs 1x
On-Demand
from $2.57
Google Cloud In stock
USA 48 configs 1x-8x
On-Demand
from $2.60
Reserved
from $1.75
Spot
from $0.27
AWS In stock
USA 6 configs 1x-8x
On-Demand
from $3.06
Reserved
from $1.30
Azure In stock
USA 6 configs 8x
On-Demand
from $1.53
Reserved
from $1.70
Spot
from $0.31
Sesterce In stock
France 4 configs 1x-4x
On-Demand
from $2.57
GPU.ai In stock
UAE 2 configs 1x-2x
On-Demand
from $0.11
HyperAI In stock
Hong Kong 4 configs 1x-8x
On-Demand
from $0.14
TensorDock
USA 1 config 1x
On-Demand
from $0.28
Geodd
USA 3 configs 1x-4x
On-Demand
from $0.16
Paperspace
USA 2 configs 1x
On-Demand
from $2.34
Oracle Cloud
USA 3 configs 1x-8x
On-Demand
from $2.95
OVH
France 3 configs 1x-4x
On-Demand
from $0.77
Heads up: We do our best to keep these specs & prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates & exact specs with the provider before provisioning.
Frequently Asked Questions
Why choose the V100?
32GB or 16GB HBM2 with 300 GB/s NVLink. Legacy data center GPU with wide availability and low cost. FP64 support makes it useful for scientific computing.
When is the V100 not a good fit?
Volta architecture lacks BF16, TF32, and FP8 support. Significantly slower than Ampere or Hopper for AI workloads. Only consider if budget is the primary constraint.
Are V100 prices going up or down?
The median on-demand price has held steady around $0.97/hr per GPU across providers.
What size AI models can the V100 run?
With 32GB of VRAM, the V100 can typically run 7B to 13B models in FP16, or larger models in 4-bit quantized form.
How much VRAM does the V100 have?
The V100 has 32GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
What is the V100's memory bandwidth?
The V100 has 900 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.
What data types does the V100 support?
The V100 supports 4 precision formats. Training: FP16, FP32. Inference: INT8. Scientific: FP64.
Does the V100 support NVLink?
Yes. The V100 supports NVLink with 300 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.
How much does the V100 cost per hour?
V100 pricing currently ranges from $0.05/hr to $4.22/hr per GPU, depending on the provider, instance type, and billing model.
How much does the V100 cost per month?
At 720 hours per month, one V100 can cost between $39.20 to $3,040.78 per month, depending on the provider. Reserved and spot pricing can lower that further.
Which cloud providers offer the V100?
The V100 is available from 21 cloud providers, including Verda, Amazon Web Services, Exoscale. Pricing and availability vary by region and billing model.
Can I rent the V100 in the cloud?
Yes. We currently track 175 V100 listings across 21 cloud providers:
Billing type Listings Avg $/GPU/hr On-demand 75 $1.57/hr Reserved 77 $1.12/hr Spot 23 $0.23/hr
Technical Specifications
| V100 PCIe | V100 SXM2 | V100S PCIe | |
|---|---|---|---|
| GPU Architecture | NVIDIA Volta | NVIDIA Volta | NVIDIA Volta |
| NVIDIA Tensor Cores | 640 | 640 | 640 |
| NVIDIA CUDA® Cores | 5,120 | 5,120 | 5,120 |
| Double-Precision Performance | 7 TFLOPS | 7.8 TFLOPS | 8.2 TFLOPS |
| Single-Precision Performance | 14 TFLOPS | 15.7 TFLOPS | 16.4 TFLOPS |
| Tensor Performance | 112 TFLOPS | 125 TFLOPS | 130 TFLOPS |
| GPU Memory | 32 GB / 16 GB HBM2 | 32 GB HBM2 | 32 GB HBM2 |
| Memory Bandwidth | 900 GB/sec | 900 GB/sec | 1134 GB/sec |
| ECC | Yes | Yes | Yes |
| Interconnect Bandwidth | 32 GB/sec | 300 GB/sec | 32 GB/sec |
| System Interface | PCIe Gen3 | NVIDIA NVLink™ | PCIe Gen3 |
| Form Factor | PCIe Full Height/Length | SXM2 | PCIe Full Height/Length |
| Max Power Consumption | 250 W | 300 W | 250 W |
| Thermal Solution | Passive | Passive | Passive |
| Compute APIs | CUDA, DirectCompute, OpenCL™, OpenACC® | CUDA, DirectCompute, OpenCL™, OpenACC® | CUDA, DirectCompute, OpenCL™, OpenACC® |
Source: official Nvidia V100 datasheet.
Alternatives to Nvidia V100
Nvidia L40S
Ada Lovelace inference GPU. From $0.20/hr per GPU across 32 providers.
Nvidia H100
Hopper datacenter GPU. From $0.60/hr per GPU across 52 providers.
Nvidia RTX 4090
Consumer Ada Lovelace GPU. From $0.14/hr per GPU across 16 providers.
Nvidia RTX 3090
Consumer Ampere gaming GPU. From $0.06/hr per GPU across 8 providers.
Nvidia T4
Turing inference GPU. From $0.06/hr per GPU across 9 providers.
Last updated