FastGPTFastGPT

Document Parse

Parse file links into text and pass the result to downstream nodes.

Function

The document-parsing node reads a list of file links and outputs text for a model or another node. Enable application file uploads or supply links accessible to the service from an upstream node. File types and size limits follow your deployment's upload and parsing configuration.

Inputs, configuration, and outputs

ItemTypeMeaning
Document LinkString arrayURLs to parse, referencing uploaded files or upstream results
Document Parsing ResultStringText for downstream nodes
Raw ResponseObject arrayOriginal per-file parsing results for further processing

Document Parse node configuration

If parsing fails, check the error in Run Details.

Purpose

Use parsing before summarization, extraction, or question answering. Knowledge-base retrieval requires its own import and indexing process; this node reads files for the current workflow.

Example: summarize an uploaded file

  1. Enable uploads in application file-input settings and prepare a short document containing readable text.
  2. Connect the start node to Document Parse and then to AI Chat.
  3. Reference the start node's file-link list as the Document Link input.
  4. Reference the Document Parsing Result in AI Chat and set instructions to Summarize the supplied document in three bullet points. Preserve dates and amounts.
  5. Upload the file in debug preview. Inspect the parsed text before checking the summary.

Document Parse debug output and summary

Compare every date and amount with the source. For empty text, inspect scans, source content, and parsing services. For reading failures, check URL reachability, expiration, and upload limits. See file input for the full capability guide.

Edit on GitHub

Last updated

On this page

Put your business knowledge to work with AI

Import company documents and build accurate, reliable, traceable AI answers.