mirror of
https://github.com/labring/FastGPT.git
synced 2026-04-27 02:08:10 +08:00
Archer
87b0bca30c
* cloud doc * doc refactor * doc move * seo * remove doc * yml * doc * fix: tsconfig * fix: tsconfig
58 lines
2.3 KiB
Plaintext
58 lines
2.3 KiB
Plaintext
---
|
|
title: V4.8.22 (Includes Migration Script)
|
|
description: FastGPT V4.8.22 Release Notes
|
|
---
|
|
|
|
## Upgrade Guide
|
|
|
|
### 1. Back up your database
|
|
|
|
### 2. Update images:
|
|
|
|
- Update the fastgpt image tag to v4.8.22
|
|
- Update the fastgpt-pro commercial edition image tag to v4.8.22
|
|
- Sandbox image update is not required
|
|
|
|
### 3. Run migration script
|
|
|
|
Only required for commercial edition users providing SaaS services.
|
|
|
|
From any terminal, send an HTTP request. Replace `{{rootkey}}` with the `rootkey` from your environment variables, and `{{host}}` with your **FastGPT domain**:
|
|
|
|
```bash
|
|
curl --location --request POST 'https://{{host}}/api/admin/initv4822' \
|
|
--header 'rootkey: {{rootkey}}' \
|
|
--header 'Content-Type: application/json'
|
|
```
|
|
|
|
This migrates contact information to the corresponding user table.
|
|
|
|
## New Features
|
|
|
|
1. AI chat node parses `<think></think>` tag content as chain-of-thought, enabling thinking process output for various models. You need to manually enable model thinking output.
|
|
2. Chat API optimization: conversation logs are now saved regardless of whether a `chatId` is provided. If no `chatId` is passed, a random one is generated for storage.
|
|
3. PPIO model provider.
|
|
|
|
## Improvements
|
|
|
|
1. Better prompts when models are not configured, reducing conflicting messages.
|
|
2. Usage record code improvements.
|
|
3. Content extraction node: long field descriptions now wrap. Output names now use `key` instead of `description`.
|
|
4. Team management interaction improvements.
|
|
5. Chat API non-stream responses now include error fields.
|
|
|
|
## Bug Fixes
|
|
|
|
1. Thinking content not counted toward output tokens.
|
|
2. Thinking chain stream output sometimes out of order with main content.
|
|
3. API workflow calls: images that don't support HEAD detection were being filtered out. Added error detection to prevent incorrect filtering.
|
|
4. Some template marketplace templates had errors.
|
|
5. Guest window unable to correctly detect whether language recognition is enabled.
|
|
6. Chat log export not compatible with sub-path deployments.
|
|
7. Member list not refreshing when switching teams.
|
|
8. List API null pointer possibility when joining member data.
|
|
9. Workflow base nodes unable to upgrade.
|
|
10. Vector search results not deduplicated.
|
|
11. User selection node unable to connect properly.
|
|
12. Chat record source not being saved correctly.
|