GoogleCloudPlatform/accelerated-platforms

Latest recorded source activity: September 29, 2026 at 18:27

SkillsMP catalog refreshed: November 9, 2026 at 14:22

skills collected: 4

GitHub stars: 107
GitHub forks: 37

Install with an AI assistant

Copy this prompt into the AI assistant you're using.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/GoogleCloudPlatform/accelerated-platforms.

Repository README

Repository README source: README.md

Read the source README

Expand full README

# Google Cloud Accelerated Platforms This repository is a collection of accelerated platform best practices, reference architectures, example use cases, reference implementations, and various other assets on Google Cloud. An accelerated platform utilizes specialized hardware components, or accelerators, such as [GPUs (Graphics Processing Units)](https://cloud.google.com/gpu) and [TPUs (Tensor Processing Units)](https://cloud.google.com/tpu), to significantly increase the speed of computationally intensive tasks. These tasks may include data analysis, machine learning, artificial intelligence, and graphics rendering. By offloading demanding workloads from traditional CPUs to dedicated hardware accelerators, which are capable of much faster parallel calculations, the platform optimizes high-performance computing. ## Cloud Workstations (CWS) > [!NOTE] > The Cloud Workstations (CWS) Platform is currently in beta and is still being > actively developed. The [Cloud Workstations (CWS) Platform](/docs/platforms/cws/README.md) is a core, best practices, fully managed workstation environments built to meet the needs of security-sensitive enterprises. It enhances the security of workstation environments while accelerating onboarding and productivity. - [Cloud Workstations reference architecture](/docs/platforms/cws/reference-architecture.md) - [Cloud Workstations reference implementation](/docs/platforms/cws/reference-implementation.md) ## Google Kubernetes Engine (GKE) ### GKE Base Platform The [GKE Base Platform](/docs/platforms/gke/base/README.md) is an implementation of a foundational platform built on GKE that incorporates best practices and provides a core environment optimized for running accelerated workloads. It offers a streamlined and efficient solution to leverage the benefits of GKE as the primary runtime. #### Example implementations - [ComfyUI reference implementation](/platforms/gke/base/use-cases/inference-ref-arch/examples/comfyui/README.md) - [Federated learning](/docs/platforms/gke/base/use-cases/federated-learning/README.md) - [Inference reference architecture](/docs/platforms/gke/base/use-cases/inference-ref-arch/README.md) - [AI inference proof-of-concept (PoC) and capacity-finding guide](/docs/platforms/gke/base/use-cases/inference-ref-arch/poc-capacity-and-rapid-deployment-guide.md) - [Inference reference implementation](/platforms/gke/base/use-cases/inference-ref-arch/terraform/README.md) - [Online inference with GPUs](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-gpu/README.md) - [Online inference using Diffusers with GPUs on Google Kubernetes Engine (GKE)](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-gpu/diffusers-with-hf-model.md) - [Online inference using vLLM with GPUs on Google Kubernetes Engine (GKE)](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-gpu/vllm-with-hf-model.md) - [Speculative Decoding using vLLM with GPUs on Google Kubernetes Engine (GKE)](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-gpu/vllm-spec-decoding-with-hf-model.md) - [Online inference with TPUs](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-tpu/README.md) - [Online inference using MaxDiffusion with TPUs on Google Kubernetes Engine (GKE)](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-tpu/max-diffusion-with-hf-model.md) - [Online inference using vLLM with TPUs on Google Kubernetes Engine (GKE)](/docs/platforms/gke/base/use-cases/inference-ref-arch/online-inference-tpu/vllm-with-hf-model.md) - Async and Batch Inference - [Async inference with GPUs](/docs/platforms/gke/base/use-cases/inference-ref-arch/async-inference/README.md) - [Batch inference with GPUs](/docs/platforms/gke/base/use-cases/inference-ref-arch/batch-inference/README.md) - llm-d - [Well-Lit paths](/docs/platforms/gke/base/use-cases/inference-ref-arch/llmd/well-lit-paths/llmd-well-lit-paths.md) - Run your own benchmarking - [Benchmarking Online inference performance on Google Kubernetes Engine (GKE)](/docs/platforms/gke/base/use-cases/inference-ref-arch/inference-perf-bench/inf-perf-benchmarking-with-hf-model.md) - [Training reference architecture](/docs/platforms/gke/base/use-cases/training-ref-arch/README.md) - [Model fine tuning](/docs/platforms/gke/base/use-cases/training-ref-arch/model-fine-tuning/README.md) - [Data processing](/docs/platforms/gke/base/use-cases/training-ref-arch/model-fine-tuning/data-processing.md) - [Data preparation](/docs/platforms/gke/base/use-cases/training-ref-arch/model-fine-tuning/data-preparation.md) - [Fine tuning](/docs/platforms/gke/base/use-cases/training-ref-arch/model-fine-tuning/fine-tuning.md) - [Model evaluation](/docs/platforms/gke/base/use-cases/training-ref-arch/model-fine-tuning/model-evaluation.md) - Supervised fine tuning - [Single-host SFT using MaxText with TPUs](/docs/platforms/gke/base/use-cases/training-ref-arch/sft-tpu-maxtext-single-host/README.md) - [Multi-host SFT using MaxText with TPUs](/docs/platforms/gke/base/use-cases/training-ref-arch/sft-tpu-maxtext-multi-host/README.md) - [Reinforcement Learning reference architecture](/docs/platforms/gke/base/use-cases/reinforcement-learning/README.md) - [Single-host reinforcement learning with TPUs using GRPO algorithm](/docs/platforms/gke/base/use-cases/reinforcement-learning/single-host-tpu-grpo/README.md) - [Secure Runtimes for AI Agents reference architecture (GKE Agent Sandbox & Agent Substrate)](/docs/platforms/gke/base/use-cases/agent-substrate-security-demo/README.md) - [Secure Runtimes for AI Agents reference implementation (gVisor sandboxing, zero-secret egress, millisecond memory snapshot restore)](/platforms/gke/base/use-cases/agent-substrate-security-demo/terraform/README.md) ### Guides - [LLM Inference Optimization: Achieving faster Pod Startup with Google Cloud Storage](/use-cases/inferencing/cost-optimization/gcsfuse/AchievingFasterPodStartup.md) - [Optimizing GKE Workloads with Custom Compute Classes](/docs/guides/optimizing-gke-workloads-with-custom-compute-classes/README.md) ### [Deprecated] Playground AI/ML Platform on GKE The [Playground AI/ML Platform on GKE](/platforms/gke-aiml/playground/README.md) is a quick-start implementation of the platform that can be used to familiarize yourself with the GKE architecture and to get an understanding of various concepts covered in the use cases. - [GKE AI/ML Platform for enabling AI/ML Ops](/docs/platforms/gke-aiml/README.md) #### Use cases - [Scalable and Distributed LLM Inference on GKE with vLLM](/docs/use-cases/inferencing/README.md) - [Retrieval Augmented Generation (RAG) pipeline](/docs/use-cases/rag-pipeline/README.md) ## Contributing For more information about contributing to this repository, see [CONTRIBUTING](/CONTRIBUTING.md).

Loading the repository README…

Skills in this repository

Network & Computer Systems Administrators Software Developers

2 occupation categories · 50% classified

skill

occupation

description

updated

llm-d-benchmarking

unclassified

Benchmarks a deployed llm-d well-lit path guide stack on GKE using llm-d-benchmark cli with inference-perf as the harness, the user a will select the appropriate workload profiles from llm-d-benchmark.

2026-09-29

llm-d-deploy-stack

unclassified

Deploys the llm-d stack on GKE using well-lit paths specification.

2026-09-29

llm-d-workload-tuner

Software Developers

This is an experimental Skill. It automatically tunes GKE vLLM inference server parameters and resources based on workload profiles specified in the benchmark configs.

2026-08-20

gke-inference-stack-deploy

Network & Computer Systems Administrators

Deploys the GKE base platform and inference-specific terra-services (GPU/TPU) for accelerated workloads.

2026-08-18

Showing 4 of 4 collected skills.