Model Serving Articles

Page 1 of 1 (4 articles)

8/2/2026 • EN

The AI Compatibility Chain: From Server Firmware to Model Runtime

Analyzes AI infrastructure upgrade risks, emphasizing compatibility chains from server firmware to model runtime for reliable GPU deployments.

AI Infrastructure Gpu Compatibility Kubernetes Model Serving Nvidia Stack

Paul Bryant

5/18/2026 • EN

Energy use of AI inference – estimates and efficiency opportunities

Analysis of AI inference energy use, comparing estimates from OpenAI, Google, and Microsoft's new framework for measuring energy under production conditions.

AI Energy Data Center Overhead Inference Optimization LLM Efficiency Model Serving

David Mytton

5/27/2024 • EN

Understanding the Cost of Generative AI Models in Production

Analyzes the complex total cost of ownership for deploying generative AI models in production, beyond just raw compute expenses.

generative ai Kubernetes Model Serving Production Deployment Total Cost Of Ownership

Philipp Schmid

12/21/2023 • EN

LLM Model Serving on Autopilot

A guide to deploying and running your own LLM on Google Kubernetes Engine (GKE) Autopilot for control, privacy, and cost management.

Autopilot Gke Kubernetes llm Model Serving

William Denniss

Model Serving Articles

The AI Compatibility Chain: From Server Firmware to Model Runtime

Energy use of AI inference – estimates and efficiency opportunities

Understanding the Cost of Generative AI Models in Production

LLM Model Serving on Autopilot

Select Language

We use cookies