|
<< Click to Display Table of Contents >> Raynet One Data Hub > 2026.2 > Connectors > Alphabetic Connector List > Generic AI Service Configuration |
The configuration section defines what data the AI receives as input, what the AI is asked to do, and how the output is structured. These settings are entered in the Configuration section of the task configuration.
Optionally select one or more existing Data Hub tables to provide as context to the AI. The contents of the selected tables are included in the request sent to the AI service. Use the search field to locate tables by name and confirm a selection to add it to the list. Added tables are shown as removable tags. To remove a table, click the delete icon next to it.
Technical Name |
sourceTables |
Type |
List of table references |
Required |
No |
The instruction or question sent to the AI service. The prompt is combined with any selected source table data and submitted to the AI provider. Write a clear and specific prompt to obtain well-structured results.
Technical Name |
promptMessage |
Type |
String (multiline) |
Required |
Yes |
Defines the output table structure that the AI is expected to populate. Each response table has a name and a set of columns. For each column, specify the column name and its data type. The AI result is written into these tables when the task runs.
Use the Add response tables button to add a new table entry and enter the table name. Expand a table row to define its columns. Use the New column button within the expanded row to add columns. For each column, provide a name and select the data type. Use the Remove button to delete a column, or select one or more table rows and click Remove table to delete entire tables.
Technical Name |
tables |
Type |
List of table definitions |
Required |
No |
The Advanced Options section provides additional parameters that control the behavior of the AI model. Expand the Advanced options accordion to access these settings.
Sets the maximum length of the AI response. Higher values allow longer answers, while lower values produce shorter and more concise replies.
Technical Name |
maxTokens |
Type |
Integer |
Required |
No |
Default Value |
4096 |
Controls the creativity and variability of the AI response. Lower values produce more focused and deterministic output. Higher values make responses more varied and creative. The valid range is 0.00 to 2.00.
Technical Name |
temperature |
Type |
Decimal (step 0.01) |
Required |
No |
Valid Range |
0.00 – 2.00 |
Default Value |
0.1 |
Limits the AI word choices to the most probable options (nucleus sampling). Lower values make responses more controlled and predictable. Higher values allow a wider variety of phrasing. The valid range is 0.00 to 1.00. It is generally not recommended to change both Temperature and Top P at the same time.
Technical Name |
topP |
Type |
Decimal (step 0.01) |
Required |
No |
Valid Range |
0.00 – 1.00 |
Default Value |
0.1 |