NEURA | IANUSTEC AI icon

NEURA | IANUSTEC AI

Interact with NEURA | IANUSTEC AI services using OpenAI-compatible API

Actions6

Chat → Create

AI-generated

Summary

Send a non-streaming chat completion request to the configured NEURA | IANUSTEC API for each input item. The node returns the first choice’s content and message metadata, with optional tool definitions and response storage settings.

Inputs

  • Model (required) — Model identifier to use; defaults to NEURA-IANUSTEC.
  • User Message (required) — Message sent with the user role.
  • System Message — Optional system instruction, trimmed and omitted when empty.
  • Temperature — Sampling randomness from 0 to 2; defaults to 0.7 and is clamped to that range.
  • Top P — Nucleus sampling value from 0 to 1; defaults to 1 and is clamped to that range.
  • Max Tokens — Positive token limit to send; an empty or non-positive value leaves the limit unset.
  • Tools JSON — Optional JSON describing tools, either an array, an object with a tools array, or a single function tool object. An object may also include tool_choice, which is passed through when tools are present.
  • Advanced Options — Store Raw Response adds a raw field containing the same formatted result in the output JSON. Binary File Name saves that formatted result as a JSON binary property with the supplied name.

Output shape

One n8n item per input item, each with a JSON object containing content, finish_reason, model, created, usage, and the full first-choice message.

Requests are non-streaming. content is an empty string if the first choice has no content; the operation errors if the API returns no choices. If Continue On Fail is enabled, failures produce JSON with error and itemIndex. The optional binary file contains the formatted result object, not the unmodified API response.

Links

Discussion