Reload the model each time ?
#20
by Steph83 - opened
Hello ! I use ComfyUI. I have Total VRAM 16376 MB, total RAM 32581 MB
pytorch version: 2.3.0+cu121
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 4070 Ti SUPER : cudaMallocAsync
Using pytorch cross attention
With so much memory, I have a waste of time at every Prompt Queue because :
“Requested to load AutoencoderKL
Loading 1 new model”
Any help ?