☁️ Google Vertex AI Login (gcloud) — REMOVED

⚠️ Vertex AI support was removed from LrGeniusAI in August 2026.

The plugin no longer offers any Vertex AI option: the Vertex AI Project ID / Vertex AI Location fields are gone from the Plug-in Manager, Analyze & Index no longer creates Vertex embeddings, and Advanced Search no longer offers the Semantic (Vertex AI) option. Vertex embeddings already stored in your database are kept, but they are neither updated nor searched anymore.

This page is kept as historical reference for the previous setup. Nothing below applies to the current version — use Choosing an AI Model to pick a supported provider instead.

If you want to use Vertex AI with LrGeniusAI, run the login on the machine where geniusai-server is running.

macOS

1. Install Google Cloud CLI (if needed):
   https://cloud.google.com/sdk/docs/install
2. Open Terminal and run:

gcloud init
gcloud config set project YOUR_PROJECT_ID
gcloud auth application-default login

3. Optional verification:

gcloud auth application-default print-access-token

Windows (PowerShell)

1. Install Google Cloud CLI (if needed):
   https://cloud.google.com/sdk/docs/install
2. Open Google Cloud SDK Shell (or PowerShell with gcloud in PATH) and run:

gcloud init
gcloud config set project YOUR_PROJECT_ID
gcloud auth application-default login

3. Optional verification:

gcloud auth application-default print-access-token

Remote backend with Docker Compose

If your backend runs as a remote Docker container, authenticate inside the container and persist the Google Cloud CLI state with the bind mount in docker-compose.yml.

1. Open a shell on the server in the repository root:

mkdir -p gcloud
docker compose up -d --build

2. Set the Vertex project inside the running container:

docker compose exec geniusai-server gcloud config set project YOUR_PROJECT_ID

3. Login for Application Default Credentials (ADC):

docker compose exec geniusai-server gcloud auth application-default login

4. Optional verification:

docker compose exec geniusai-server gcloud auth application-default print-access-token

For headless SSH hosts without a browser, use:

docker compose exec geniusai-server gcloud auth application-default login --no-browser

Then follow the remote bootstrap flow shown by gcloud on a second trusted machine that has a browser and Google Cloud CLI installed.

Notes

• gcloud auth application-default login creates local Application Default Credentials (ADC).
• In Docker Compose, the bind mount ./gcloud:/root/.config/gcloud keeps ADC and the active gcloud project across container restarts and rebuilds.
• Set Vertex AI Project ID and Vertex AI Location in the Lightroom plugin settings.
• Do not set GOOGLE_APPLICATION_CREDENTIALS when you want the container to use ADC created by gcloud auth application-default login.
• For headless/server deployments, prefer service-account auth via GOOGLE_APPLICATION_CREDENTIALS.