AreChen
AreChen
你让他总结然后复制单条对话
3080ti的话是不是int4都跑不了,内存有64G,我把上面大佬提供的代码加在moss_inference.py下会提示OOM ``` Traceback (most recent call last): File "moss_inference.py", line 352, in model = MossForCausalLM.from_pretrained("fnlp/moss-moon-003-sft-plugin-int4").half().cuda() File "/home/arechen/.conda/envs/moss/lib/python3.8/site-packages/torch/nn/modules/module.py", line 680, in cuda return self._apply(lambda t: t.cuda(device)) File "/home/arechen/.conda/envs/moss/lib/python3.8/site-packages/torch/nn/modules/module.py", line 570, in...
Okay, I will try it later. Thank you!
I also hope to have this function. You can configure the API yourself to use the AI function, similar to the [novel project](https://github.com/steven-tey/novel).
需求+1,现在AWS bedrock 有claude3,使用非常方便,希望可以添加这个功能 Requirement +1, now AWS bedrock has claude3, which is very easy to use. Hope this feature can be added. 
同问这个插件无法安装
DigitalHumanAnimation就是METAHUMAN SDK，去下这个插件自己替换
另外这个插件在中文区不提供需要切换英文才能下载到