TensorFlow Serving

AWS Azure Application Stacks 2 variations

TensorFlow Serving for production model inference, enforced behind nginx basic auth with per instance credentials on first boot.

Base
Hardened build
minimal ports, security patches applied at build time
Access
Unique credentials
generated on first boot, readable only by root
Verified
Boots working
services pass a health gate before release
Support
24/7, 365 days
by email and live chat, 24 hour response SLA

Variations

Choose the operating system, cloud and version you need. Every variation is the same hardened TensorFlow Serving image, built and supported by cloudimg.

ProductCloudOperating systemVersion
TensorFlow Serving With Nginx Auth AWS Standard View · Guide
TensorFlow Serving 2 Ubuntu 24.04 Azure Ubuntu 24.04 2 View · Guide

Overview

TensorFlow Serving is a high performance serving system for machine learning models, exposing trained SavedModels through REST and gRPC inference APIs. It is designed to serve model predictions in production and to manage model versions over time.

Why the cloudimg image

The cloudimg image is hardened and fully patched with TensorFlow Serving running under Docker Compose and systemd, a sample model preloaded and nginx basic auth enforced on every request. Per instance credentials are generated on first boot and stored in a root only file so no endpoint is exposed without authentication, and 24/7 cloudimg support is included.

Common uses

  • Model inference serving
  • Model version management
  • Production machine learning