GPUDatacenterReleased Nov 26, 2019

NVIDIA Tesla V100S PCIe 32GB

4096-bit · 5,120 cores

The Tesla V100S PCIe 32GB is a faster refresh of NVIDIA's Volta-based Tesla V100 PCIe data center accelerator, announced in November 2019. It keeps the same GV100 GPU and 250 W power rating but raises clock speeds and memory bandwidth, delivering about 16-17% more compute performance and 26% more bandwidth than the original V100 PCIe. It targets AI training, inference and double-precision HPC workloads in standard PCIe servers.

General

Manufacturer: NVIDIA
Model: Tesla V100S PCIe 32GB
Segment: Datacenter
GPU chip: GV100
Architecture: Volta
Release date: Nov 26, 2019
Price: Search on Amazon

Memory

Memory size: 32 GB
Memory type: HBM2
Bus width: 4096-bit
Bandwidth: 1,134 GB/s
Memory clock: 1,107 MHz

Clocks & silicon

GPU clock: 1,245 MHz
Boost clock: 1,597 MHz
Shader units: 5,120 CUDA Cores
Tensor / AI cores: 640
RT cores: —
Transistors: 21.1 billion
Technology (nm): 12 nm
TSMC 12 nm FFN
TDP / TBP: 250 W

Interface & APIs

Bus interface: PCIe 3.0 x16
SLI / multi-GPU: None (no NVLink on PCIe version)
DirectX: N/A (compute-only)
OpenGL: N/A (compute-only)
Vulkan: N/A (compute-only)
Other APIs: CUDA 7.0, OpenCL 3.0

AI support

Precisions: FP8FP16BF16INT8INT4TF32FP64
AI accelerator: 1st-gen Tensor Cores (640)
Peak AI TOPS: —
Notes: The 640 first-generation Tensor Cores perform FP16 matrix multiplication with FP32 accumulation for up to 130 TFLOPS of deep-learning throughput. FP64 runs at half the FP32 rate, 8.2 TFLOPS versus 16.4 TFLOPS. BF16, TF32 and FP8 arrived only with later architectures, and INT8 is handled by the DP4A instruction on the CUDA cores.

All NVIDIA GPUs →

Datacenter gpus from NVIDIA announced around the same time as the NVIDIA Tesla V100S PCIe 32GB. Add any of them to your compare list to see the specs side by side.

gpu