deepak-newzera

How to generate longer samples using pre-trained model

1

I am using the pre-trained model from https://huggingface.co/riffusion/riffusion-model-v1 as follows for generating music from a text prompt: model_id = "riffusion/riffusion-model-v1" pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16) I am getting the music of...

I am running the model on a batch of 500 articles and it takes around 10 to 15 GB of GPU memory. Is there a way to optimize the GPU...