run-openai-model

Description

The run-openai-model adaptor uses OpenAI Chat Completions to create a model response for a given chat message.

Note that this adaptor can be used to create a single response for one chat message only. If you would like to create a response for each value in a datatable column, use the transform-columns adaptor. If the values are in a list, you can use the convert-list-to-datatable adaptor to convert the list to a datatable before running the model.

Inputs

model Type: text Required: No The name of the language models to run. Consider the price points of each model. If unspecified, defaults to gpt-3.5-turbo.

system message Type: text Required: Yes The system message helps set the behavior of the assistant (e.g., You are a helpful assistant). You can modify the personality of the assistant or provide specific instructions about how it should behave throughout the conversation.

user message Type: text Required: Yes The input of the model to run.

api token Type: text Required: Yes OpenAI API token

Outputs

assistant message Type: text The model prediction.

duration Type: text The duration of the model run.

Examples

Example 1: Find the capital of a country.

Inputs:

model: gpt-3.5-turbo

system message: You will be provided with a country name and your task is to return the name of its national capital.

user message: United Kingdom

api token: sk-************************************************ (visit API Keys page to retrieve the API key)

Outputs:

assistant message: London

duration: 1019