Actions16
- LLM (Text Generation) Actions
- Image Generation Actions
- Video Generation Actions
- Text-to-Speech Actions
- Speech-to-Text Actions
- Music Generation Actions
- Embeddings Actions
- Content Moderation Actions
- Custom Inference Actions
LLM (Text Generation) → Complete
AI-generatedSummary
Generate text completions from a prompt using a specified Chutes.ai large language model chute. Customize generation parameters such as temperature, max tokens, penalties, stop sequences, and response format.
Inputs
- Chute — Select a specific chute URL to use for text generation or use a dynamic expression to specify a chute URL. Defaults to the standard LLM chute at Chutes.ai.
- Prompt (required) — The text prompt to generate completions from.
- Additional Options — Optional parameters to control generation randomness (temperature), maximum tokens to generate, nucleus sampling via top_p, frequency and presence penalties to reduce repetition, stop sequences to end generation early, response format (text or JSON), streaming toggle, random seed for deterministic output, and optional timeout in seconds for the API call.
Output shape
A single record containing the generated text completion or JSON response.
The output depends on the selected response format option ('text' returns plain text, 'json_object' returns a JSON object). The node performs a single API call to the Chutes.ai LLM chat completions endpoint using the prompt wrapped as a user message. Streaming is disabled by default. If a timeout is specified and exceeded, the node errors and can trigger n8n's retry flow.