This document details the technical improvements released in FastGPT 4.14.7 for self-hosted deployments, focusing on usability, stability, compatibility, and error reduction across core platform features.
Core User Experience & Tooling Updates
This section addresses front-end usability and tool call reliability:
- Enhanced user experience for tool selection and Dataset selection within the Chat Agent interface, streamlining configuration for chatbot builders.
- Tool calls now automatically populate empty arguments with
"{}"to avoid runtime errors from model providers that do not support empty string values. - Added native compatibility for Kimi 2.5 tool calls when the model operates in thinking mode, aligning platform support with the latest model capabilities.
- Improved internal network domain validation to restrict access to authorized resources and reduce security risks.
Backend Stability & Compatibility Fixes
These changes reinforce platform reliability and compatibility with standard infrastructure:
- The MCP component automatically filters out extraneous fields during save operations, maintaining full compatibility with MongoDB 4.x deployments.
- The backend layer now automatically filters unconfigured tools to prevent model errors caused by attempts to call undeployed or unconfigured tools. A shared filter function is used across both frontend and backend to ensure consistent behavior between the two layers.
- Orphaned edges in workflow diagrams are automatically removed prior to workflow execution, eliminating broken triggers that would cause execution failures.
Workflow and API Configuration Enhancements
This section covers targeted improvements to workflow and API functionality, with key changes summarized in the table below:
| Improvement Area | Technical Change |
|---|---|
| Workflow AI Model Setup | Added memory selection support for workflow AI models when operating in chat log mode |
| API File Link Processing | When calling workflows via API with file links, save file type directly from input instead of inferring from URL, ensuring 100% accurate file type detection |
Source: FastGPT official source
Applicability and version scope
Use this page for the documented Deployment and upgrades scenario. Confirm the FastGPT, dependency, API, and deployment versions in the official source before applying a change.
Safety guardrails
Use [REDACTED_CREDENTIAL] for credentials and private data. Confirm the documented environment and version before review.
Rollback guidance
Restore the prior technical-content authority snapshot. Restore saved configuration and data snapshots, then repeat the smallest verification scenario.