In this step by step docs, you will see how to use AI Inside Google Sheets, to run multiple variations of your prompts. In total it will take 8-10 min to get started then it will be as simple as opening any google sheet and running the function
Get started with SheetAI AppGet StartedInstall the add-on (1 min)Create OpenAI Account and get API key (4 min)Setup API key to SheetAI App (2 min)SheetAI ExamplesSHEETAI functionSHEETAI_LIST functionSHEETAI_TABLE functionSHEETAI_EDIT functionSHEETAI_TAG functionSHEETAI_EXTRACT functionSHEETAI_TRANSLATE functionSHEETAI functionSHEETAI_FILL functionGeneral Functions Settings
- Go to the installation page
- Click on the Install button
SheetAI App - Google Workspace Marketplace
Run variations of GPT3 prompts right inside google sheets, to Sanitise Data, Write personalised Thank You cards, Summarise Feedback, Categorise Feedback and more
- Click on Continue
- Pick your Google account
- Allow the requested permissions
- And you’re done!
To create OpenAI key we need to SignUp for OpenAI account and then verify with mobile number and then generate a key.
Creating OpenAI Account does not cost money, Initially you get $18 in credits which is enough to write 1000’s of 3000 letter blogs.
- Create an account at https://platform.openai.com/signup
- Click Create new secret key
- Copy it to your clipboard
Just visit sheet.new (to create new google sheet) and then Extensions > SheetAI App > Launch
SheetAI makes automating mundane tasks like text editing, generating multiple answers, and extracting information easily. With SheetAI, you can quickly and easily create taglines, apply tags, translate text, summarise, and populate data with dummy information.
Here is a List of SheetAI Functions with usage Examples.
Write a tagline:
Add some temperature for more creativity:
If you want the AI response to be listed, take up multiple cells in a Google Sheet.
Like GPT, but outputs the results in a table. Very practical when the output is a table.
Edit your text:
Apply all relevant tags:
Extract email addresses or country:
Translate from English to Hindi:
Summarise a text:
Populate descriptions in a product catalog:
Clean up an email list, Clean up names, Generate dummy data and more.
you can set
- response_token (which helps set the token limit for response initially it is 200 if need a larger output increase it. For English language, 1000 tokens are 750 words (source),
- temperature: Controls Randomness, Lowering results in less Random Completion. As the temperature approaches zero model become deterministic and repetitive
- mode: you can use all models “gpt-3.5-turbo”, “text-curie-001”, “text-babbage-001”, “text-ada-001”
- cache: helps save openai credits in case of rerun of SheetAI function, it is saved in your secure google cache for 6 hours.