ethanfel
ethanfel
> > You have to tune the batch_size to fit your VRAM and the load-thread to fit your RAM. > > A smaller batch_size will have a negative impact on...
> > you use nvidia-smi and htop to look at your vram and ram to tune your parameters > > ``` > Total : 8192 MiB > Used : 6721...
With gpu ? export CUDA_DEVICE_ORDER=PCI_BUS_ID parameter : --gpu_id 0,1,2,3,..
i can confirm. Moving to 3.10.6 fix the issue for me
enable the upscaler
It's working great
> The ControlNet extension improves the control over the composition by a lot. I would imagine it being a worthwhile addition. > > **Additional context** Currently the extension even throws...
my only issue is the fact that it won't reconnect between session. Seems like the ssh key is refused by the container after running the colab script and ./autorisation. I...
> Cannot verify. [comfyanonymous/ComfyUI@a28a9dc](https://github.com/comfyanonymous/ComfyUI/commit/a28a9dc83684624ee2167c0b92d976bb68f2c606) seems to work fine. Added a simple WF to illustrate the issue. start comfyui -> run -> error. but bypass layer diff -> run (work) ->...
the bug is still present in the later comfyui commit.