Thanks akhaliq, I got a community gpu for this space. Can I transfer it to the blocks organization and keep the gpu allocation?

@simonduerr , we are working on a fix for load gradio demo from spaces so it can be easy as import gradio as gr gr.Interface.load("simonduerr/
protGPT2_gradioFold").launch() in the gradio-blocks organization to create a copy of this space and use the same gpu allocation. For now can you try the api workaround for 3.0 spaces here https://huggingface.co/spaces/versae/gradio-blocks-rest-api and let me know if this works

If I try to use the API i get this error. The function takes about 200 sec to compute.
curl -X POST https://hf.space/embed/simonduerr/protGPT2_gradioFold/+/api/predict/ -H 'Content-Type: application/json' -d '{"data": ["MPSYEIPEFLKTTDNIRRISPSSVLLRFLEKFLNDLQEEDGTVVRHPPIEVRLPIERDVAIGPSLRFFSLIKKVKEVKGYSLDFVRREYDEFLLWDFLDRLREKSGNNVLEVLRIENCKR"], "fn_index": 1}'