Deployment and upgradesOfficial documentation6 min read版本解读

FastGPT V4.6.5 New Features and Fixes

This article records configuration and changes from a historical FastGPT release.

This article records configuration and changes from a historical FastGPT release. Use it to understand that release, and consult the release notes for your installed and target versions before applying dependencies, configuration, or migration steps.

New Workflow Nodes

This release adds four dedicated workflow automation nodes with official documentation guides:

  1. Query Rewriting Node: Refines user input queries to enhance retrieval accuracy and response relevance within workflow pipelines
  2. Text Editor Node: Supports in-line text manipulation, formatting, and content transformation for intermediate workflow steps
  3. Classifier Node: Routes workflow execution paths based on predefined content classification criteria
  4. Custom Feedback Node: Enables structured user feedback collection to iterate on workflow output quality

Enhanced Content Extraction Node

The existing Content Extraction node receives two critical functional updates, summarized in the parameter table below:

ParameterNew Supported Capability
Model SelectionAllows selection of a dedicated model for content extraction tasks, replacing the fixed default model
Field EnumerationsPermits definition of explicit output fields to structure extracted content into predefined, standardized categories

Parsing and Performance Optimizations

Two key infrastructure and document parsing improvements are included:

  1. DOCX document parsing now supports conversion of embedded tables to Markdown format, preserving structured table data from source DOCX files with greater accuracy
  2. Resolved a CPU-intensive thread blocking issue caused by the html2md processing pipeline, eliminating unexpected resource contention during HTML-to-Markdown conversion workflows

Bug Resolutions

This release addresses two key workflow interface and functionality issues:

  1. Fixed inaccurate or incomplete prompt extraction descriptions within the advanced orchestration interface, correcting context gaps for prompt-based workflow steps
  2. Improved advanced orchestration connection line interactions to streamline visual workflow design and reduce user friction when building complex pipelines

Source: FastGPT official documentation and source