Skip to content
Updated minutes ago
nvidia

V100 16GB

nvidia · volta · 16 GB HBM2 · 250W TDP

VRAM

16 GB

BF16 TFLOPS

28.3

Bandwidth

900 GB/s

From

$0.15/hr

Calculate ROI with this GPU →

Spec Sheet

VRAM16 GB HBM2
Memory Bandwidth900 GB/s
BF16 TFLOPS28.3
FP16 TFLOPS28.3
FP8 TFLOPS28.3
INT8 TOPS56.5
TDP250W
InterconnectNVLINK
NVLink Bandwidth300 GB/s
Max per Node8
PCIe Gen3
CUDA Compute Capability7
Tensor CoresYes

Pricing by Provider

ProviderOn-DemandReservedSpotBadge
vast_ai$0.29/hr-$0.15/hrCheapest
tensordock$0.25/hr-$0.15/hr
runpod$0.39/hr-$0.25/hr
aws$3.06/hr$1.96/hr$0.92/hr
gcp$2.48/hr$1.58/hr-

Compatible Models (246)

Training Capabilities

Estimated GPU count for full fine-tuning (AdamW, BF16) and QLoRA

Model SizeFull Fine-TuneQLoRA
7B model9 GPUs1 GPU
13B model16 GPUs1 GPU
70B model83 GPUs3 GPUs

Energy Efficiency

Estimated tokens/second per Watt for popular models

Mistral 7B
0.49 t/s/WFP8
Qwen 2.5 7B
0.47 t/s/WFP8
Llama 3.1 8B
0.45 t/s/WFP8
Llama 3.1 70B
0.05 t/s/WFP8
Qwen 2.5 72B
0.05 t/s/WFP8

Similar GPUs

GPUVRAMBF16 TFLOPSBW (GB/s)From
V100 32GB32 GB28.3900$0.19/hr
A400016 GB76448$0.17/hr
RTX 408016 GB97717$0.32/hr
T416 GB65300$0.12/hr
RTX 4060 Ti 16GB16 GB44288$0.30/hr

Methodology Note

Performance estimates for the V100 16GBare based on InferenceBench's roofline performance model with CUDA kernel-level optimization including FlashAttention v2 and PagedAttention. Memory calculations account for model weights (16 GB HBM2 available), KV-cache allocation, and activation memory. Throughput predictions use the V100 16GB's rated 900 GB/s memory bandwidth and 28.3 BF16 TFLOPS compute capacity as roofline ceilings, with empirical correction factors per GPU architecture (volta). See our full methodology.

Frequently Asked Questions

How many AI models can run on V100 16GB?

The V100 16GB can run 246 AI models from our database within a single node. Compatible models range across various parameter sizes depending on the quantization precision (BF16, FP8, INT4). Smaller models fit on a single GPU while larger models may require multi-GPU setups up to 8x V100 16GB.

What is the V100 16GB inference throughput?

The V100 16GB delivers 28.3 BF16 TFLOPS and 28.3 FP8 TFLOPS with 900 GB/s memory bandwidth. Actual inference throughput (tokens/sec) depends on the model size, precision, and batch size. Use our calculator for model-specific throughput estimates.

How much does V100 16GB cost per hour?

The V100 16GB is available starting from $0.15/hour via vast_ai. Prices vary by provider and pricing tier (on-demand, reserved, spot). Compare pricing across all providers in the table above.