AI hardware
Google TPU v2
FP16 throughput
4.6×10¹³ FLOP/s
Memory
14.9 GB
TDP
280 W
Google TPU v2 is an AI accelerator designed by Google (TPU), released in May 2017. It is built on a 16 nm process.
It delivers 4.6×10¹³ FLOP/s of tensor FP16/BF16 throughput. It carries 14.9 GB of memory at 559 GB/s of bandwidth. Its thermal design power is 280 W.
Full record
- Manufacturer
- Type
- TPU
- Release date
- May 1, 2017
- Tensor FP16/BF16
- 4.6×10¹³ FLOP/s
- FP32
- 3×10¹² FLOP/s
- Memory
- 14.9 GB
- Memory bandwidth
- 559 GB
- TDP
- 280 W
- Process node
- 16 nm
More from Google
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