could not control computer

[zhao0112]

Could you please elaborate on why the instructions do not continue with the subsequent steps the user should take? I'm curious to understand the reasoning behind the decision to conclude the instructions at that point, as I'd be very interested in learning the full scope of how the VLMPlanner application is intended to be used. I'm using qwen2vl + ShowUI

[h-siyuan]

From the screenshot provided, it seems that the model is running, could you provide more context?

1. how is the memory usage after the VLMPlanner message is printed?

2. What are the CPU and memory size of your Mac?

[zhao0112]

My Mac is equipped with an M1 Pro chip and 16GB of RAM.After the VLMPlanner message is printed, the memory usage reaches 14GB out of 16GB of RAM.Do I need to enable any additional system permissions?

[h-siyuan]

It might be the speed of model inference being slow, we have provided a guide for int8 quantization, you can refer to https://github.com/showlab/ShowUI/blob/main/QUICK_START.md