LDXhubInstall

Summary

Uploads the input file to LDXhub, runs a RefineLoop job with the selected model and settings, polls until completion, and downloads the refined file.

Inputs

• Input Binary Field (required) — Binary property containing the file to refine; defaults to data.
• Model Name or ID (required) — LDXhub AI model to use. Choose a loaded model or provide its ID.
• Max Revisions — Maximum refinement iterations (1–10); defaults to 6.
• Options — Optional custom instructions, subject domain, numeric-segment exclusion, note language, output mode (full, translations, or none), and hyphenation removal. Defaults include full output mode, hyphenation removal enabled, and numeric-segment exclusion disabled.
• Polling Settings — Controls job polling: maximum attempts defaults to 180, and server wait defaults to 10 seconds (1–10).

Output shape

One output item per input item, containing job metadata in JSON and the downloaded refined file as binary under the same property name as the input file.

JSON includes job_id, status, output_file_id, and usage. The file is downloaded only when the completed job returns an output file ID. Errors are thrown unless the node is configured to continue on fail, in which case an error item is returned.

Examples

Example 1: Workflow use

Connect a node that provides the source file as binary data, set its binary property name and an LDXhub model, then optionally adjust refinement and polling settings.

Links

• n8n community nodes installation guide