Google data center Available

Google TPU v5e

TPU v5 ยท v5e Architecture

Enquire Now
The Google TPU v5e is a cost-efficient AI accelerator designed for inference and fine-tuning workloads. With 16GB HBM and 197 TFLOPS BF16 per chip, it offers the best price-performance ratio in Google Cloud's TPU lineup. Ideal for serving large language models, generating images, and running medium-scale training jobs.

Key Features

Cost-efficient design 16GB HBM per chip Optimized for inference Scalable pod configurations JAX/TensorFlow support

Full Specifications

Compute

Architecture TPU v5e
BF16 Performance 197 TFLOPS

Memory

Memory Size 16 GB
Memory Type HBM
Memory Bandwidth 819 GB/s

Power & Physical

Form Factor Custom ASIC

Features & Connectivity

NVLink Support No
Multi-GPU Support Yes

Availability

MSRP (USD) Contact for pricing
Release Date Aug 2023
Status Available

Industries

AI/ML Cloud Computing Research

Use Cases

AI Inference Model Fine-tuning Text Generation Image Generation Embedding Models

Interested in the Google TPU v5e?

Get pricing, availability, and bulk discount information from our team.

Enquire Now

Related GPUs

Google data center

Google TPU v5p

Memory

95GB HBM

Available View Specs
Google data center

Google TPU v4

Memory

32GB HBM

Available View Specs