Try Duet AI in the Google Cloud console
Learn how to ask for assistance by chatting with Duet AI in the Google Cloud console.
Duet AI doesn't use your prompts or its responses as data to train its model. For more information, see How Duet AI in Google Cloud uses your data.
Before you begin
Open Duet AI
In the Google Cloud console, go to any page—for example, the Dashboard page.
In the console toolbar, select a Google Cloud project that's associated with a project ID you submitted after you were granted access to Duet AI.
The chat_spark Duet AI icon appears in the toolbar.
Click chat_spark Duet AI.
The Duet AI pane opens.
If necessary, click Accept if you agree to the terms.
The pane displays a few example prompts that you can click, or you can enter a prompt in the Enter a prompt here field. The displayed prompts vary depending on your current Google Cloud console page.
Chat with Duet AI
In the following steps, you learn to enter a prompt along with a follow-up question:
In the Enter a prompt here field, enter the following prompt, and then click send Send:
Create a gcloud command to give the developer Google group access to view my Google Cloud project.
Duet AI displays its answer after the prompt.
In the Enter a prompt here field, enter a follow-up prompt, and then click send Send:
Change that to editor access.
Duet AI uses your conversational history for additional context when it responds to your prompts. For example, Duet AI responds with an updated explanation and command.
If more information is available, the link Show related content appears after the response. To access the information, click Show related content.
To expand Duet AI to the full width of your browser window, click open_in_new in the Duet AI pane. The Duet AI pane opens your current session in a new tab.
More examples to try
In the following examples, you learn how to get information about various Google Cloud products by chatting with Duet AI in the Duet AI pane:
To learn about the differences between similar products, enter a prompt that is similar to the following:
What are some of the pros and cons to using Anthos over GKE for my application?
To learn how to get started with a new Google Cloud project, enter a prompt that is similar to the following:
I'm interested in developing a computer vision application. I have a lot of unlabelled training data sitting in my Google drive and I'm not sure how to get started. What are the first steps I should take to develop an image recognition service with this data?
This prompt returns step-by-step information about how to create a Google Cloud project.
To learn about a log entry in Logs Explorer, you can click a button on a log entry that opens Duet AI chat with a bulleted summary of the events in the log. For more information, see Summarize log entries with Duet AI assistance.
For information, see Write better prompts for Duet AI.
Reset chat history
Duet AI resets the chat history automatically when you change projects or sign out of the Google Cloud console.
If your chat history is no longer relevant to what you're currently trying to do, then you can reset the chat history:
- In the Duet AI pane, click the icon, and then click Reset Chat.
Because your chat history is available only in the browser that you use to access Google Cloud and is not stored elsewhere, you can also remove your chat history by doing one of the following:
Close your browser or the tab in your browser that you're using to access Google Cloud.
Refresh the tab you're using to access Google Cloud.
If you minimize your chat conversation by clicking the underscore
_) on the Duet AI pane, your chat history isn't removed.
Save chat history
- To save your chat history, copy your conversation into a document, and then save the document.
Use Duet AI chat outside the console
To use Duet AI in an integrated development environment (IDE) instead of the Google Cloud console, you must first install the Cloud Code plugin in your IDE.
- To chat with Duet AI in your IDE, click chat_spark Duet AI.
Chatting with Duet AI in an IDE works the same way as chatting the Google Cloud console.