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.
Key Changes in FastGPT Commercial Edition 4.6.6
This document outlines the mandatory configuration adjustments for self-hosted FastGPT Commercial Edition when upgrading to version 4.6.6. Three core changes apply to all deployments: updating the commercial edition container image, relocating a critical plugin configuration to an environment variable, and removing a legacy configuration section from the config.json file.
Step-by-Step Configuration Migration
Follow these exact steps to complete the configuration update without disruption:
- Update your FastGPT Commercial Edition container image to the 4.6.6 release.
- Locate the
SystemParams.pluginBaseUrlentry in your existingconfig.jsonfile and remove it entirely, as this setting is no longer read from the configuration file. - Configure the new
PRO_URLenvironment variable with your commercial edition plugin service address. This variable replaces the formerSystemParams.pluginBaseUrlsetting. Note that the address must not end with/api. A valid example is:PRO_URL=http://fastgpt-plugin.ns-hsss5d.svc.cluster.local:3000. Add this variable to your deployment configuration, such as a Docker run command, Kubernetes manifest, or.envfile. - Delete the entire
FeConfigsection from yourconfig.jsonfile, as this section has been fully removed and is no longer supported. - Restart your FastGPT Commercial Edition services to apply all configuration changes.
Centralized Dashboard Configuration
All platform parameters and model configurations that were previously managed via the FeConfig section or other config.json entries can now be adjusted directly through the FastGPT Commercial Edition web dashboard. Users are no longer required to edit the config.json file for standard platform setup or parameter tweaks, reducing manual configuration errors and simplifying future maintenance.