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
Custom Inference → Batch
AI-generatedSummary
Send a batch of inputs for custom inference to a specified deployed model on Chutes.ai, optionally using a selected chute URL, with configurable job priority, output format, webhook callback URL, and timeout.
Inputs
- Chute — Select a specific chute URL to use for the custom inference request; defaults to https://llm.chutes.ai if not specified.
- Model ID (required) — The unique identifier of the deployed model on Chutes.ai to run inference against.
- Batch Inputs (required) — A JSON array containing multiple input data objects for batch processing by the model.
- Additional Options — Optional settings including maximum timeout in seconds for the inference call (errors if exceeded), a webhook URL for asynchronous results, job priority level (low, normal, high), and desired output format (JSON or raw).
Output shape
a single JSON response object containing individual inference results for each batch input
The batch request is sent to the /v1/inference/{modelId}/batch endpoint of the selected chute. If a timeout is set and exceeded, the node errors and can trigger retries. When webhookUrl is provided, results may be delivered asynchronously to that URL. Output format controls the structure of returned data (parsed JSON or raw). The response includes inference results mapped to each input in the batch.