Parallel AIInstall

Summary

Submits a chat conversation to the selected Parallel AI model and returns its completion. Provide the required model ID and messages, with optional sampling, penalty, response format, tool, and user settings.

Inputs

• Model (required) — Model ID to use for the completion.
• Messages (required) — JSON array of conversation messages.
• Temperature — Sampling temperature (default 1; declared range 0–2).
• Top P — Nucleus sampling parameter (default 1; declared range 0–1).
• Max Tokens — Maximum tokens to generate (default 0).
• Frequency Penalty — Frequency penalty from -2.0 to 2.0 (default 0).
• Presence Penalty — Presence penalty from -2.0 to 2.0 (default 0).
• N — Number of completions (default 1; the UI says only 1 is supported).
• Logit Bias — JSON mapping of token IDs to bias values.
• Response Format — JSON response format specification.
• Seed — Random seed for reproducibility (default 0).
• Stop — Stop sequence or sequences.
• Stream — Whether to stream the response (default false).
• Tool Choice — Tool choice specification.
• Tools — JSON tool definitions.
• User — End-user identifier.

Output shape

A completion response from the chat request.

The provided source pack contains no executable implementation body, so the precise returned data shape and runtime handling of optional parameters cannot be confirmed from the source.

Links

• n8n community nodes installation guide