AI & LLM
Lido provides action formulas for calling large language models directly from your spreadsheet. Each formula sends a prompt to an AI service and writes the response to a specified output cell.
GPT
Calls an OpenAI GPT model.
Syntax:
=GPT(credential, prompts, output_ref)
=GPT(credential, prompts, output_ref, model, max_tokens, response_format, temperature, usage_output, tools)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | OpenAI credential ID | Yes |
prompts | Prompt text or array of messages | Yes |
output_ref | Cell reference to write the response | Yes |
model | Model name (e.g., "gpt-4o") | No |
max_tokens | Maximum tokens in the response | No |
response_format | Response format (e.g., "json") | No |
temperature | Sampling temperature | No |
usage_output | Cell reference for token usage stats | No |
tools | Tool/function definitions for function calling | No |
GPTVISION
Analyzes images or documents using GPT Vision.
Syntax:
=GPTVISION(files, credential, prompt, output_ref)
=GPTVISION(files, credential, prompt, output_ref, page_ranges, model, max_tokens, response_format, image_detail, usage_output)
Parameters:
| Parameter | Description | Required |
|---|---|---|
files | File URL(s) to analyze | Yes |
credential | OpenAI credential ID | Yes |
prompt | Instructions for the vision model | Yes |
output_ref | Cell reference to write the response | Yes |
page_ranges | Page ranges for PDFs (e.g., "1-3") | No |
model | Model name | No |
max_tokens | Maximum tokens in the response | No |
response_format | Response format | No |
image_detail | Image detail level ("low", "high", "auto") | No |
usage_output | Cell reference for token usage stats | No |
CLAUDE
Calls an Anthropic Claude model.
Syntax:
=CLAUDE(credential, prompt, output_ref)
=CLAUDE(credential, prompt, output_ref, model, max_tokens, system_message, files, page_ranges, usage_output)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | Anthropic credential ID | Yes |
prompt | Prompt text | Yes |
output_ref | Cell reference to write the response | Yes |
model | Model name (e.g., "claude-sonnet-4-20250514") | No |
max_tokens | Maximum tokens in the response | No |
system_message | System message for the model | No |
files | File URL(s) to include | No |
page_ranges | Page ranges for PDFs | No |
usage_output | Cell reference for token usage stats | No |
BEDROCK
Calls an AWS Bedrock model.
Syntax:
=BEDROCK(credential, prompt, output_ref)
=BEDROCK(credential, prompt, output_ref, model, max_tokens, system_message, files, page_ranges, usage_output)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | AWS access key credential ID | Yes |
prompt | Prompt text | Yes |
output_ref | Cell reference to write the response | Yes |
model | Model name | No |
max_tokens | Maximum tokens in the response | No |
system_message | System message for the model | No |
files | File URL(s) to include | No |
page_ranges | Page ranges for PDFs | No |
usage_output | Cell reference for token usage stats | No |
GEMINI
Calls a Google Gemini model.
Syntax:
=GEMINI(credential, files, prompt, output_ref)
=GEMINI(credential, files, prompt, output_ref, page_ranges, model, max_tokens, response_format, response_schema, usage_output)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | Google API key credential ID | Yes |
files | File URL(s) to analyze | Yes |
prompt | Prompt text | Yes |
output_ref | Cell reference to write the response | Yes |
page_ranges | Page ranges for PDFs | No |
model | Model name | No |
max_tokens | Maximum tokens in the response | No |
response_format | Response format | No |
response_schema | JSON schema for structured output | No |
usage_output | Cell reference for token usage stats | No |
GPTURL
Analyzes the content of a URL using GPT.
Syntax:
=GPTURL(credential, url, prompt, output_ref)
=GPTURL(credential, url, prompt, output_ref, model, max_tokens)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | OpenAI credential ID | Yes |
url | URL to fetch and analyze | Yes |
prompt | Instructions for the model | Yes |
output_ref | Cell reference to write the response | Yes |
model | Model name | No |
max_tokens | Maximum tokens in the response | No |
EXTRACTTOTABLE
Uses AI to extract structured data from content into a table.
Syntax:
=EXTRACTTOTABLE(credential, input_content, output_table_ref)
=EXTRACTTOTABLE(credential, input_content, output_table_ref, prompt, model, max_tokens, usage_output)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | AI credential ID | Yes |
input_content | Content to extract data from | Yes |
output_table_ref | Cell reference for the output table | Yes |
prompt | Custom extraction instructions | No |
model | Model name | No |
max_tokens | Maximum tokens | No |
usage_output | Cell reference for token usage stats | No |
EXTRACTFROMFILETOROW
Uses AI to extract data from a file and write it to a row.
Syntax:
=EXTRACTFROMFILETOROW(credential, file_source, output_ref)
=EXTRACTFROMFILETOROW(credential, file_source, output_ref, prompt, model, max_tokens, usage_output)
Parameters:
| Parameter | Description | Required |
|---|---|---|
credential | AI credential ID | Yes |
file_source | File URL to extract data from | Yes |
output_ref | Cell reference for the output | Yes |
prompt | Custom extraction instructions | No |
model | Model name | No |
max_tokens | Maximum tokens | No |
usage_output | Cell reference for token usage stats | No |