The SEND_REQUEST action is used to send a request to the OpenAI large language model (LLM), such as data summarisation, content enrichment or suggestions for the end user input.
The input parameters to send an OpenAI request are:
|
Parameter |
Type |
Required |
Description |
|---|---|---|---|
| model | String | Required | Name of the OpenAI model to use. Refer to models compatible with the /v1/chat/completions API as defined in the Model endpoint compatibility of the OpenAI Platform documentation. For example: gpt-4o. |
| prompt | String | Required | The prompt that will be sent to the OpenAI model. For example: Explain the concept of gravity. |
| systemInstructions | String | Optional | Additional instructions for the OpenAI system. For example: The answer must be in the HTML format. |
| temperature | String | Optional | The temperature that will be used by the OpenAI system to generate a response. Higher
values will make the output more random, lower values will make it more
deterministic. The value of this parameter must be within a range
between 0 and 2. The default value is 0.7. Note: It is recommended to use either
Temperature or Top
P parameter, not both at the same
time.
|
| topP | String | Optional | The top P value that will be used by the OpenAI system to generate a response. It is an
alternative to sampling with temperature. The value of this parameter
must be within a range between 0 1. The default value is
1. Note: It is recommended to
use either Temperature or Top
P parameter, not both at the same
time.
|
| maxCompletionTokens | Integer | Optional | The maximum number of tokens to generate in the completion. The value of this parameter must be within a range between 1 and 4096. For example: 100. |
The output parameters to send an OpenAI request are:
|
Parameter |
Type |
Required |
Description |
|---|---|---|---|
| response | String | Optional | A response from the OpenAI model. |