STACKIT Chat Model icon

STACKIT Chat Model

Provides access to the STACKIT Chat Model API.

STACKIT Chat Model

AI-generated

Summary

Generate text completions or chat-based responses using a specified STACKIT Chat Model with configurable generation options.

Inputs

  • Model (required) — Select the model to use for generating the completion. The list is dynamically retrieved from the /models endpoint, excluding models containing 'intfloat'.
  • Options — Optionally configure generation parameters such as frequency penalty, presence penalty, max tokens, temperature, top P, max retries, timeout in milliseconds, and select response format (text or JSON).
  • If using JSON response format, you must include word "json" in the prompt in your chain or agent. — Notice shown only when JSON response format is selected, ensuring the prompt includes the word 'json' to generate valid JSON output.

Output shape

a single JSON object containing the generated response, output under property 'response'

Output is returned under the 'response' property in the 'ai_languageModel' output. For JSON responseFormat, the generated content will be valid JSON; otherwise, plain text is returned. Errors during execution are output as NodeErrors. Retries and timeout settings control request robustness.

Discussion