Skip to content

Latest commit

 

History

History
85 lines (49 loc) · 3.52 KB

google-vertexai.md

File metadata and controls

85 lines (49 loc) · 3.52 KB

Google VertexAI

Prerequisites

  1. Start your GCP
  2. Install the Google Cloud CLI

Setup

Enable vertex AI API

  1. Go to Vertex AI on GCP and click "ENABLE ALL RECOMMENDED API"

Create credential file (Optional)

There are 2 ways to create credential file

No. 1 : Use GCP CLI

  1. Open terminal and run the following command
gcloud auth application-default login
  1. Login to your GCP account
  2. Check your credential file. You can find your credential file in ~/.config/gcloud/application_default_credentials.json

No. 2 : Use GCP console

  1. Go to GCP console and click "CREATE CREDENTIALS"

  1. Create service account

  1. Fill in the form of Service account details and click "CREATE AND CONTINUE"
  2. Select proper role (for example Vertex AI User) and click "DONE"

  1. Click service account that you created and click "ADD KEY" -> "Create new key"

  1. Select JSON and click "CREATE" then you can download your credential file

Flowise

Without credential file

If you are using a GCP service like Cloud Run, or if you have installed default credentials on your local machine, you do not need to set this credential.

With credential file

  1. Go to Credential page on Flowise and click "Add credential"
  2. Click Google Vertex Auth

  1. Register your credential file. There are 2 ways to register your credential file.

  • Option 1 : Enter path of your credential file
    • If you have credential file on your machine, you can enter the path of your credential file into Google Application Credential File Path
  • Option 2 : Paste text of your credential file
    • Or you can copy all text in the credential file and paste it into Google Credential JSON Object
  1. Finally, click "Add" button.
  2. 🎉You can now use ChatGoogleVertexAI with the credential in Flowise now!

Resources