It takes 1 to 2 minutes for the model to create the video

[lahfir]

I've deployed the model on runpod with the specified requirements. A100 with 80GB specs. But still, it takes 1 to 2 minute to produce the video even for a 5 seconds audio. How can I solve this?

[daarkwulf]

how u deployed can u tell.. coz here it takes 1.5 hours for 8 sec audio

[lahfir]

@AIhasArrived https://replicate.com/cjwbw/sadtalker

[AIhasArrived]

What about huggingface, is it paid ? Do you think it can be less expensive to use it instead? I have more faith on hugigng face as it is better known

[AIhasArrived]

@lahfir

[lahfir]

For hugging face, you'll be paying for deploying and hosting your models. But in order to use the already hosted models, you'll be just paying for the usage

[AIhasArrived]

have you tried deployed there? Is it espensive? I wonder in the end if it wil be better to have a model there

[AIhasArrived]

Did you guys find a way to make it faster @lahfir @daarkwulf please answer.

[lahfir]

#540 (reply in thread)

[lahfir]

It's all about usage, my friend. The payment is based on how much you use it. It's a pay-as-you-go system.

[AIhasArrived]

you replied to the wrong mini discussion but its okay:) , can you point me to sad talker hugging face api? It is probably les expensive than replicate no?