NVIDIA V100

Specifications

V100 supported

• FP16 vector
• FP16 matrix
• FP32 vector
• FP64 vector
• INT8 vector

Performance

The following are theoretical maximum performance in TFLOPS:¹

Data Type	Vector	Matrix	Sparse
FP64	7.8
FP32	15.7
TF32	15.7
FP16	31.4	125
BF16
FP8
INT32
INT8

Footnotes

1. https://images.nvidia.com/aem-dam/en-zz/Solutions/data-center/nvidia-ampere-architecture-whitepaper.pdf ↩