I am running Flux 1 dev text to image model through ComfyUI in Kaggle. Everything works but I noticed that Kaggle offers a second GPU inside the notebook. If I try to run two instances of the ComfyUI and run the two identical workflows in parallel the notebook runs out of memory.
My question is: Is there a way to use both GPUs for inference, without breaking the RAM memory limit, given that both flows are identical?
My final objective is to achieve inference with double speed (since both GPUs will be working) or at least speed up the process with the second GPU, if this is possible.