WatsonX LLM icon

WatsonX LLM

For advanced usage with an AI chain and tracked via Langfuse

WatsonX LLM

AI-generated

Summary

Provides a Watsonx chat model for connection to an AI chain. Select a foundation or custom model, set the required API version, and optionally tune generation settings or request JSON output.

Inputs

  • Model Name or ID (required) — Choose an available foundation or custom model, or specify its ID using an expression.
  • API Version (required) — Watsonx API version; defaults to 2024-05-31.
  • Options — Optional generation settings: maximum new tokens (default 1024), temperature (0–2, default 0.7), top P (0–1, default 1), and output format. JSON modes enable native JSON output; “JSON Mode with Prompt Injection” also appends an instruction to supported prompt inputs to return only a valid JSON object.

Output shape

An AI Language Model connection for an AI chain; this operation does not return regular workflow items.

Requires Watsonx API credentials. JSON Mode Only does not add a prompt instruction, so the prompt should request JSON output.

Links

Discussion