city96 / ComfyUI_ExtraModels

Support for miscellaneous image models. Currently supports: DiT, PixArt, HunYuanDiT, MiaoBi, and a few VAEs.
Apache License 2.0
394 stars 35 forks source link

`/models/T5` should be `/models/t5` #55

Closed liusida closed 6 months ago

liusida commented 6 months ago

in the README.md, you mentioned:

Download the [second text encoder from here](https://huggingface.co/Tencent-Hunyuan/HunyuanDiT/blob/main/t2i/mt5/pytorch_model.bin) and place it in ComfyUI/models/T5 - rename it to "mT5.bin"

But when i install that on Linux, the node is looking for the ComfyUI/models/t5 folder. So i think T should be lower case.

Thanks.

city96 commented 6 months ago

You're right, it's lowercase everywhere else, even in the code where the folder creation happens. Updated the readme to reflect this.

I've also added a link to a decoder-only fp16 safetensor version of mT5-xl that I converted, which should be smaller/load faster compared to the pytorch model.