SL LLMs icon

SL LLMs

StrangeLogic LLM Node
Join our community

Actions2

Overview

This node integrates with StrangeLogic's Large Language Models (LLMs) to generate text or process video content based on user input. It is useful for scenarios where automated text generation or video analysis is needed, such as creating content, answering questions, or extracting information from videos. For example, users can input a prompt to generate creative writing or provide a video URL for processing and analysis.

Use Case Examples

  1. Generating a text response from a prompt using a specified LLM model.
  2. Processing a video URL to extract insights or perform analysis using the video processing operation.

Properties

Name Meaning
Model Name or ID Select the LLM model to use for text generation or video processing.
Prompt The input text prompt for the LLM to respond to.
Temperature Controls randomness in the generated text, affecting creativity and variability.
Max Tokens Maximum length of the generated text response.
Top P Nucleus sampling parameter controlling the diversity of token selection.
Top K Limits token selection to the top K tokens for generation.
Saftey Settings : Hate Block Safety filter level to block hate speech content (applicable to certain models).
Saftey Settings : Harrasment Block Safety filter level to block harassment content (applicable to certain models).
Saftey Settings : Sexual Block Safety filter level to block sexual content (applicable to certain models).
Saftey Settings : Dangerous Content Block Safety filter level to block dangerous content (applicable to certain models).
JSON Response Option to receive the response in JSON format instead of plain text (available for specific models).

Output

JSON

  • json - The response from the LLM or video processing API, containing generated text or processed video data.

Dependencies

  • Requires an API key credential for StrangeLogic AI API access.

Troubleshooting

  • Ensure valid API credentials are provided; missing credentials will cause errors.
  • If the API response is not an array when loading models, the node will throw an error indicating invalid API response.
  • Network or API errors during request will throw errors unless 'Continue On Fail' is enabled, in which case errors are returned in the output JSON.

Links

Discussion