nele.ai icon

nele.ai

Search for results in your nele.ai knowledge database
Join our community

Actions4

Overview

This node integrates with the nele.ai platform to generate AI-powered chat completions based on user-provided messages. It is designed to interact with a conversational AI model, allowing users to send a sequence of messages (including roles like user, assistant, and system) and receive generated text responses. This node is useful for automating customer support chats, creating interactive assistants, or generating conversational content dynamically.

A practical example would be feeding a conversation history to the node and receiving a context-aware reply from the AI model, which can then be used in chatbots or virtual assistants.

Properties

Name Meaning
Model Name or ID The AI model that will generate the text response. Users can select from a predefined list or specify a custom model ID using expressions.
Messages A collection of messages forming the conversation history. Each message includes:
- Role: Defines the speaker role, options are User, Assistant, or System.
- Content: The actual text content of the message.
Knowledge Database Name or ID (Optional) If set, the node retrieves data from this knowledge database to provide contextual information relevant to the last user message. Can be selected from a list or specified via expression.
Max Tokens Maximum number of tokens allowed in the generated reply. Controls the length of the AI's response.
Temperature Sampling temperature controlling randomness of the output. Values range from 0 (more deterministic) to 1 (more random).

Output

The node outputs JSON data containing the AI-generated completion text based on the input messages and parameters. The main output field json includes the generated response text from the AI model.

If the node supports binary data output (not explicitly shown here), it would typically represent audio or image data related to the AI response, but for the Chat Completion operation, the output is primarily textual.

Dependencies

  • Requires an API key credential for authentication with the nele.ai service.
  • Depends on the external nele.ai API endpoint configured via environment variables.
  • The node uses dynamic loading methods to fetch available models and knowledge databases from the service.

Troubleshooting

  • Common issues:
    • Invalid or missing API key: Ensure the API key credential is correctly configured.
    • Incorrect model or knowledge database ID: Verify that the selected model and database exist and are accessible.
    • Exceeding max tokens limit: Adjust the Max Tokens property if the response is truncated or errors occur.
  • Error messages:
    • Authentication errors usually indicate invalid credentials; recheck API key setup.
    • Validation errors may arise if required fields like messages or model are empty.
    • Network or timeout errors suggest connectivity issues with the nele.ai API endpoint.

Links and References

Discussion