Version Upgrades/4.15.x
V4.15.0-beta3
FastGPT V4.15.0-beta3 Release Notes
📦 Upgrade Guide
Image Changes
- Update the fastgpt-app (FastGPT main service) image tag to v4.15.0-beta3.
- Update the fastgpt-pro (FastGPT commercial edition) image tag to v4.15.0-beta3.
- Update the fastgpt-code-sandbox image tag to v4.15.0-beta3.
Environment Variable Changes
Code Sandbox adds security-related environment variables such as SANDBOX_API_MAX_BODY_MB and SANDBOX_MAX_OUTPUT_MB, and now supports grouped request queuing for run APIs through queueId. The full defaults are listed below:
| Variable | Default | Description |
|---|---|---|
SANDBOX_API_MAX_BODY_MB | 8 | Maximum /sandbox API JSON body size, including variables, in MB. |
SANDBOX_MAX_OUTPUT_MB | 10 | Maximum output JSON size for one code execution, including return values and logs, in MB. |
CHECK_INTERNAL_IP | true | Enables internal IP checks for sandbox network requests by default to reduce SSRF risk. |
SANDBOX_MAX_TIMEOUT | 60000 | Timeout for one code execution, in milliseconds. |
SANDBOX_MAX_MEMORY_MB | 256 | Memory limit for one sandbox, in MB. The runtime reserves an extra 50 MB for overhead. |
SANDBOX_POOL_SIZE | 20 | Number of pre-warmed JS/Python workers. |
SANDBOX_REQUEST_MAX_COUNT | 30 | Maximum number of network requests allowed during one code execution. |
SANDBOX_REQUEST_TIMEOUT | 60000 | Timeout for one network request from inside the sandbox, in milliseconds. |
SANDBOX_REQUEST_MAX_RESPONSE_MB | 10 | Maximum response body size for one sandbox network request, in MB. |
SANDBOX_REQUEST_MAX_BODY_MB | 5 | Maximum request body size for one sandbox network request, in MB. |
SANDBOX_QUEUE_ID_CONCURRENCY | Empty | Number of requests with the same queueId that may enter execution at once. Empty disables queueing. |
🚀 New Features
- Multimodal models now support audio and video input.
- Shared links and portal pages now support language switching, and no longer force browser-language auto switching.
⚙️ Improvements
- Improved styles for Skill module dialogs.
- Improved Skill list API performance.
- Improved workflow node name and description inputs.
- Workflow editor drafts are now saved automatically for recovery when the session expires and the user is redirected.
- Improved the login page UI.
🐛 Bug Fixes
- Adapted TTS audio playback to the latest OpenAI SDK to avoid errors.
- Fixed cases where Knowledge Base data chunking could produce oversized chunks when code blocks were present.
🛠️ Code Improvements
- Updated the token calculation dependency to improve performance.
- Rewrote dialog-related code with more modular structure.
- Improved unit test performance, reducing full runs from about 10 minutes to 5 minutes.
- Upgraded to TypeScript 6.
- Improved GitHub Actions security.