haoqiangyu

Refer to https://github.com/comfyanonymous/ComfyUI/issues/940 to use --disable-cuda-malloc

百度网盘：https://pan.baidu.com/s/1ESyW6ep722tD1PrkAhXPiA?pwd=5ys5  提取码:5ys5

> 使用Windows一键启动包 ，填写了 Pexels API Key 报错， 麻烦您帮忙看看哪里不对。翻墙，不翻墙都已试过，还是报错。 >  >  > 少了个方括号

@SingingDalong You can use llm on different platforms by modifying `llm_provider` and filling in the corresponding API key. 

Google Drive: https://drive.google.com/file/d/1FWUHRLrRFh1HmiMJl_3cLjjibt0hA9AL/view?usp=sharing @yapthatsme

@1060404996 https://zhuanlan.zhihu.com/p/686161924 系统问题，参考下这些解决方案

@yclwk 应该是权限问题 参考 https://blog.csdn.net/shaylo/article/details/125719378

> Here we go: [#1606 (comment)](https://github.com/huggingface/diffusers/issues/1606#issuecomment-1346667389) ;-) I don't think this guarantees the thread safety of the pipeline

> > Here we go: [#1606 (comment)](https://github.com/huggingface/diffusers/issues/1606#issuecomment-1346667389) ;-) > > I don't think this guarantees the thread safety of the pipeline If you use pytorch2.0.0 and above, it will cause...

 The lower version of the Nvidia driver cannot use cudaMallocAsync. It is also necessary to check the driver version before using it.