CLōD icon

CLōD

Interact with CLōD API (OpenAI-compatible)

CLōD

AI-generated

Summary

Send a chat completion request to the CLōD API using a specified model and an array of message objects, optionally customizing generation parameters.

Inputs

  • Model (required) — The model to use for generating the chat completion, e.g., 'Llama 3.1 8B'.
  • Messages (required) — A JSON array of message objects specifying the conversation history, each with a 'role' and 'content'.
  • Options — Optional settings to customize the response: 'Temperature' controls randomness (0-2), 'Max Tokens' limits response length, and 'Stream' enables response streaming (partial support).

Output shape

a list of output items each containing a JSON object with the API response for the corresponding input item

Each output item corresponds to an input item; in case of an error and if 'Continue On Fail' is enabled, the output contains an error message object instead.

Examples

Example 1: Using the 'Llama 3.1 8B' model to respond to a user message with customized temperature and maximum tokens.

Set 'Model' to 'Llama 3.1 8B', 'Messages' to the conversation JSON array, and under 'Options' set 'Temperature' to 0.7 and 'Max Tokens' to 512.

Links

Discussion