Quantized-t5 models on Cuda

Hello!
Are there any plans on implementing quantized-t5 models on CUDA devices?
I'm looking for a couple of days to find the solution or implement a CUDA support for https://github.com/huggingface/candle/blob/main/candle-examples/examples/quantized-t5/main.rs but I couldn't figure it out.
Would be great to be able to run the customized and quantized-t5 models with GPU support.

Start with candle-examples/examples/quantized-t5/main.rs and trace how quantized T5 execution currently selects its device. Compare that path with the repository's CUDA support, then determine the missing integration needed for GPU execution. Done means customized and quantized T5 models run successfully on CUDA devices.

Written by the indexing model from the issue text.