Inferbase
Beta
NVIDIA V100 PCIe 32GB
Datacenter
The V100 PCIe brings Volta architecture to standard servers with 32GB HBM2 and NVLink support. A reliable option for organizations with existing PCIe infrastructure.
VRAM
32 GB
Memory
HBM2
Bandwidth
900 GB/s
TDP
250W
Suitable Workloads
Smaller Language Models
Inference for 7B-13B parameter models
Enterprise Deployment
Designed for 24/7 datacenter operations
Key Highlights
- Built on Volta architecture
- 640 AI accelerator cores for tensor operations
- 5,120 compute cores for parallel processing
- PCIe form factor
Memory
VRAM32 GB
Memory TypeHBM2
Memory Bandwidth900 GB/s
Compute Performance
FP32 (Single Precision)14 TFLOPS
FP16 (Half Precision)112 TFLOPS
Architecture
ArchitectureVolta
Compute Cores5,120
AI Accelerators640
Release Year2018
Power & Physical
TDP250 W
Max Power250 W
Form FactorPCIe
PCIe GenerationGen3 x16
Interconnect
Multi-GPU SupportYes
Interconnect Bandwidth300 GB/s
Pricing
MSRP$6,500
Notes
PCIe variant of V100
Models That May Fit on NVIDIA V100 PCIe 32GB
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
Related Tools
Added Jan 25, 2026
Last updated: Jan 25, 2026
Build Your AI Stack with Confidence
Explore models, compare pricing and benchmarks, and right-size your infrastructure — all in one place.