04.6.13 AI Text Generation
All nodes in the group AI: Text Generation contain in their name:
- The name of the model to be used when the node starts;
- Prompt (preview) - an indication that the node supports simple request sending;
- With History (preview) - an indication that the node supports not only sending a request, but also sending JSON with the content of the previous interaction to improve the quality of the response to the request.
For each model, a pair of nodes Prompt (preview) and With History (preview) with appropriate settings is provided.
Node Configuration
Required and optional fields are required to customize the nodes.
- Required fields for Prompt (preview) nodes include the User Prompt field.
- Required fields for With History (preview) nodes include the User Prompt field and the Dialogue History field.
User Prompt
A text field to be filled in. The text can be filled in manually or be the result of previous nodes.
Dialogue History
Field for inserting a JSON object displaying the history of previous interaction with AI. The history can be used by the AI to generate a response in the desired format and get additional information or to receive instructions. Depending on the role, the content can be used in different ways:
- system - System messages define the identity of the AI. You can use them to set AI rules or behavior.
- user - user messages with requests.
- assistant - assistant messages prompt the AI for the desired output format.
[
{"role":"system","content":"you are a very funny"},
{"role":"user","content":"What is the capital of Australia?"},
{"role": "assistant", "content": "The capital of Australia is Canberra."}
]
Max Tokens (Answer Size)
The maximum number of response tokens that can be generated by AI before stopping. Models can stop before reaching this maximum. This parameter sets only the absolute maximum number of tokens that can be generated.
The AI: Text Generation group of nodes includes models: