Google Cloud Blog

Protecting customers with generative AI indemnification

Google Cloud assumes responsibility for potential legal risks of using our generative AI, offering indemnities for training data and generated output.

Canary analysis: Lessons learned and best practices from Google and Waze

Waze, maker of the community-based traffic navigation app, estimates that canary releases can prevent a quarter of all incidents on their services.

TPU v4 provides exaFLOPS-scale ML with efficiency gains

A new paper describes how Google’s Cloud TPU v4 outperforms TPU v3 by 2.1x on a per-chip basis, and improves performance/Watt by 2.7x.

Google Cloud Workstations managed development environment is now GA

Cloud Workstations, now GA, enables faster developer onboarding and increased productivity with an enhanced security posture.

Google Cloud TPU Multislice Training

We used Multislice Training to run the world’s largest LLM distributed training job on a compute cluster of 50,944 Cloud TPU v5e chips.

Dataflow/Beam and Spark: A Programming Model Comparison

Dataflow is a fully managed streaming analytics service that reduces latency, processing time, cost through autoscaling and real-time data processing.

AlloyDB for PostgreSQL under the hood: Columnar engine

In this technical deep dive, we take a look at the columnar engine that delivers industry leading query performance for AlloyDB for PostgreSQL.

Busting myths about BigQuery

Train TensorFlow models faster and at lower cost on Cloud TPU Pods

TPUs provide a faster and more cost-effective way to train machine learning models. Learn more about the benefits of training models on Cloud TPU Pods.

Squawk bots: Can generative AI lead us to understanding animals?

The Earth Species Project is using large language models to develop tools to decipher animal communication.

Cloud Spanner now with a free trial

Create a 90-day Spanner free trial instance with 10GB storage at no cost. Try Spanner free.

New Compute Engine A2 VMs–First Nvidia Ampere A100 GPUs in the Cloud

Google Cloud’s new Accelerator-Optimized (A2) VM family is based on the NVIDIA Ampere A100 GPU, and designed for demanding HPC and ML workloads.

GCP and Confluent partner to deliver a managed Apache Kafka service

Localllm lets you develop gen AI apps on local CPUs

Want to use open-source LLM models from Hugging Face on your local development environment? With localllm and Cloud Workstations, you can.

VM Manager: Operate large Compute Engine fleets with ease

The new VM Manager simplifies infrastructure and compliance management for the largest of Compute Engine VM fleets.

Persistent Disk performance overview | Compute Engine

Spanner: TrueTime and external consistency

Understand Spanner's TrueTime and external consistency, ensuring globally consistent reads and writes across the distributed database.

Measuring the environmental impact of AI inference

A methodology for measuring the energy, emissions, and water impact of Gemini prompts shines a light on the environmental impact of AI inference.