Contact Information
fdfbest@163.com
MaxKB Version
v2.9.0
Problem Description
AI输出时，前边输出很慢，然后突然一大段或全部内容，全部出现。
Steps to Reproduce
问题：你好
回复：2s内流式输出“Here's a thinking process:
Analyze User Input:
User said: "你好" (Hello)
There's an empty list [] at the end, which might be a placeholder or artifact, but the core message is a greeting.”
突然，一大段，或全部内容呈现。
总感觉是页面stream流速度小于AI模型输出的速度。
链接的本地AI大模型 Qwen3.6-27B-AWQ-Int4模型，模型通过CherryStudio测试速度为73 tokens /s
The expected correct result
No response
Related log output
Additional Information
No response
Contact Information
fdfbest@163.com
MaxKB Version
v2.9.0
Problem Description
AI输出时，前边输出很慢，然后突然一大段或全部内容，全部出现。
Steps to Reproduce
问题：你好
回复：2s内流式输出“Here's a thinking process:
Analyze User Input:
User said: "你好" (Hello)
There's an empty list [] at the end, which might be a placeholder or artifact, but the core message is a greeting.”
突然，一大段，或全部内容呈现。
总感觉是页面stream流速度小于AI模型输出的速度。
链接的本地AI大模型 Qwen3.6-27B-AWQ-Int4模型，模型通过CherryStudio测试速度为73 tokens /s
The expected correct result
No response
Related log output
Additional Information
No response