AI hardware
NVIDIA V100
FP16 throughput
1.3×10¹⁴ FLOP/s
TDP
300 W
NVIDIA V100 is an AI accelerator designed by NVIDIA (GPU), released in June 2017. It is built on a 12 nm process at TSMC, packing 21.1B transistors.
It delivers 1.3×10¹⁴ FLOP/s of tensor FP16/BF16 throughput. Its thermal design power is 300 W.
Full record
- Manufacturer
- NVIDIA
- Type
- GPU
- Release date
- Jun 21, 2017
- Tensor FP16/BF16
- 1.3×10¹⁴ FLOP/s
- FP32
- 1.6×10¹³ FLOP/s
- FP64
- 7.8×10¹² FLOP/s
- Memory bandwidth
- 835 GB
- TDP
- 300 W
- Process node
- 12 nm
- Transistors (millions)
- 21,100
- Die size (mm2)
- 815
- Foundry
- TSMC
- Tensor cores
- 640
More from NVIDIA
SourceEpoch AI, 'Machine Learning Hardware'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/machine-learning-hardware. Licensed under CC BY 4.0.
← All ai hardware