circulus

안녕하세요~ 좋은 연구결과를 공유해 주셔서 감사드립니다. 다름이 아니라, 현재 예제가 KoGPT2 기존 버전(v1) 및 Transformer 3 버전대에 최적화되어 있는데, 혹시 KoGPT2 v2 (Transformer 4.x) 기준으로 업데이트 예정은 없으신지 궁금합니다.

audioldm-m-text-ft ckpt to huggingface model

1

Could you have any plan to upload hf model of audioldm-m-text-ft?

spacing problem.

4

Thanks to your nice works! but I met some problem as spacing each token. for example... ... 'on' 'st' 'amps' 'and' 'sh' 'ipping' '.' ... stamps is one word and...

### Describe the issue It seems to support cuda 12 at py3.9 but 3.10 and 3.11 seems to be cuda 11.8 ### Urgency _No response_ ### Target platform ubuntu2204 ###...

I am appreciate for this application :) In the app, that is avaliable sending a voice message (like as bluetooth speaker). Could you add that function for sending a voice...

ip-adapter plus is not working with full face

6

Hi. I am trying to use ip-adapter with openvino, two model is working well (ip-adapter_sd15.bin, ip-adapter_sd15_light.bin) but, I wish to use "ip-adapter-full-face_sd15.bin", it is not working well some modification of...

I checked sana is available using optimum-intel (export). but I could not using sana-sprint which is extreamly fast. (https://huggingface.co/Efficient-Large-Model/Sana_Sprint_1.6B_1024px_diffusers) and also need to support sana1.5 Could you add this model?

[Feature] Support SANA 1.5

1

do you have any plan to add sana 1.5 model? or can i get a script for converting those model? https://huggingface.co/Efficient-Large-Model/SANA1.5_4.8B_1024px_diffusers

Sana-Sprint is so fast thought 1~4 steps for generating image that means suitable for using image-generation on device-ai Could you plan to add this task?