Icon of Make Table AI Function: Three-column gray table with white plus icon in the blue circle at the bottom right angle.

MAKE TABLE

Create and fill a table for a specific theme using ChatGPT in your spreadsheet

It quickly generates tables in your spreadsheet based on the specific theme or topic you provide. Ideal for organizing data, brainstorming sessions, or project planning, this function uses ChatGPT to create structured and insightful tables. Creating tables with sample data in your Google Sheets and then filling them with your data can significantly speed up your workflow.

Examples
=MAKETABLE_X("Fictional productds table with 25 rows")

Expected Output:
Table of products with 25 rows
=MAKETABLE_X("Fictional Book Inventory", "Title, Author, Genre, Published Year")

Expected Output:
Table of books with columns Title, Author, Genre, and Published Year
=MAKETABLE_X("Fitness Routine", "Day, Exercise ,Duration, Calories Burned", {"with_web",TRUE; "temperature",0.4; "act_like","Fitness Trainer"; "extra_instructions","Factor in rest days"})

Expected Output:
Table with a fitness schedule
=MAKETABLE_X("Our clients who are working in tech", "ID, Name, Job, Email, Phone", {"with_kb",TRUE; "temperature", IFS($C$6="Very Creative", 0.6, $C$6="Medium",0.3, $C$6="Very Rational",0)})

Expected Output:
Filtered by work table from a spreadsheet Knowledge Base
Arguments List
=MAKETABLE_X(subject, [header],[options])
=MAKETABLE_Y(subject, [header],[options])
subject
The subject specifies the theme or topic around which the table will be generated, guiding the ChatGPT in structuring the data appropriately.
For example: "Fictional table of clients with realistic data", A2, or "Table with four columns about " & B2
[optional] header
A  list of the table column names separated by commas to create the spreadsheet table with those exact columns. If not provided, ChatGPT will automatically fill in the columns.
For example: "ID, Name, Email, Phone", A2, or TEXTJOIN(", ", TRUE, A1:A5)
Options List
Particular object at the end of a formula which contains optional parameters values in the form {"parameter_name",value; "next_parameter_name",value}
For Example: {"output_language","French"; "extra_instructions","Remember this info: " & A1; "act_like","Data Analyst"}
with_kb
When activated, the function utilizes the spreadsheet's Knowledge Base to create a table, which is beneficial when you require a data-related table in your Google Sheets.
Options: FALSE or TRUE
with_kb2
When activated, the Smart Function uses Knowledge Base sources to create tables at a later stage. This is particularly helpful when used together with extra_instructions and the with_thinking options enabled.
Options: FALSE or TRUE
with_web
When activated, it enables the function to fetch data from the internet, providing real-time information from a Google search to ChatGPT. It can improve the created table but provoke extra costs and execution time.
Options: FALSE or TRUE
with_thinking
When activated, it adds an extra thinking step to improve the table. This makes the execution longer and more expensive.
Options: FALSE or TRUE
output_language
Set the output language for the table created by ChatGPT; it can differ from the language of the arguments. If not specified, the language of the answer may vary based on the options and parameter values provided.
For Example: "EN", "French", or  "Most Popular Language in the US"
temperature
This changes how creative the AI is when making the table. A creativity level higher than 0.5 might ruin the result.
For Example: 0, 0.2, or 1
act_like
A role, job, or viewpoint for the AI to use when creating a table; this setting helps the ChatGPT give more detailed and relevant results.
For example: "Products Expert", "Geographer", or "Google Sheet Table"
extra_instructions
Provide additional information or instructions to get a better table for your Google spreadsheets.
For example: "Use metric system", "Table should have 20 rows", or "Remember this: " & B2
hide_header
When set to TRUE, this option removes the header row from the resulting table, which is helpful if your spreadsheet table already includes a header. Using TEXTJOIN to provide a header row as an argument can enhance the structure of your Google Sheets. In such cases, activating this option is necessary.
Options: FALSE or TRUE
tags
Tags are used to filter Knowledge Base sources for table creation, improving the quality of the results. Initially, add tags to the Knowledge Base sheet. Then, activate with_kb or with_kb2 to enable the functionality of these tags.
For example: "employees", "products", or  B2
model
Model affects quality, speed, and tokens count of tokens to use. We choose the optimal model ("default") by default, and here it's gpt-4. Information about models.
Options: "default", "gpt-4, "gpt-3.5-turbo", or "gpt-3.5-turbo-16k"
max_tokens
Limit the number of tokens that can be used per function execution. By default, this is set to a reasonable value to prevent unwanted depletion of your balance.
1 token ≈ 4 chars in English
1 token ≈ 3/4 words
100 tokens ≈ 75 words
1-2 sentences ≈ 30 tokens
1 paragraph ≈ 100 tokens
1,500 words ≈ 2048 tokens
For Example: 100, 350, or 2048
generate
When set to TRUE, the formula creates the table. If set to FALSE, a warning message will be displayed instead of the table. To avoid accidental costs, link this parameter to a Google Sheets cell with a checkbox or a formula that verifies the input.
For Example: TRUE, FALSE, or $H$1
generate_warning
This text shows when 'generate' is set to FALSE. It's recommended to provide guidance on the steps required to initiate AI table generation. Adding the codeword [STOP] halts the processing of associated formulas, remaining invisible to the user. Typing 'default' triggers a standard warning message. 
For Example: "default", "Fill out all inputs", or $G5