Google TPU v3 32GB

TPU 2018 Spec confidence: Official
VRAM
32 GB
HBM2
Bandwidth
900 GB/s
memory
Download datasheet

Performance

Peak theoretical throughput by precision type

BF16
Brain Float 16
123 TFLOPS

Specifications

Architecture

TPU

Launch Year

2018

Memory

32 GB HBM2

Bandwidth

900 GB/s

Spec Confidence

Official

Datasheet & Resources

Flopper Datasheet

Google TPU v3 32GB specifications, generated from our database. Printable.

Vendor product page

Google documentation for the TPU v3 32GB

View

Data Provenance

Every figure traced to a source

Primary Source

Document
--
Publisher
Google
Published
--

Data Quality

Spec confidence
Official
Clock basis
Boost
Precisions tracked
1
Normalization
All values in TFLOPS

Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.

Browse GPU Cloud Providers

Similar GPUs

Frequently Asked Questions

How many TFLOPS does the Google TPU v3 have?

The Google TPU v3 delivers N/A TFLOPS for FP32 operations, N/A TFLOPS for FP16, and N/A TFLOPS for FP8 precision.

What is the power consumption of the Google TPU v3?

The Google TPU v3 has a TDP (Thermal Design Power) rating of N/A watts.

How much memory does the Google TPU v3 have?

The Google TPU v3 is equipped with 32 GB of VRAM with 900 GB/s memory bandwidth.

What architecture is the Google TPU v3 based on?

The Google TPU v3 is based on the TPU architecture, launched in 2018.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs