Claude Pro LM icon

Claude Pro LM

Language Model node for Claude via setup-token authentication.

Claude Pro LM

AI-generated

Summary

Generate text completions or responses using a selected Claude language model with customizable parameters such as extended thinking, temperature, and token limits.

Inputs

  • Model (required) — Select the Claude model to use; the list is dynamically loaded from the API and includes model-specific maximum token support.
  • Extended Thinking — Toggle to enable chain-of-thought reasoning which uses a dedicated thinking budget of tokens.
  • Thinking Budget — Set the maximum tokens allocated for the extended thinking process, active only if extended thinking is enabled (min 1024, max 128000).
  • Options — Additional generation settings including temperature (controls randomness), topP (nucleus sampling), topK (restricts token choices), max tokens override, timeout in milliseconds for the API response, and max retries on transient errors.

Output shape

a single AI-generated text response with optional tool call data

Returns the AI's generated text and may include structured tool call information if the model invokes any tools. The node includes automatic retry on transient errors and respects timeout settings. Streamed or chunked output is internally handled but the final output is the full generation response.

Links

Discussion