super slow on 3080ti,

[leafiy]

F:\ComfyUI\models\insightface\models\antelopev2\scrfd_10g_bnkps.onnx detection [1, 3, '?', '?'] 127.5 128.0
set det-size: (640, 640)
Some weights of the model checkpoint were not used when initializing CLIPTextModel:
['text_model.embeddings.position_ids']
Some weights of the model checkpoint were not used when initializing CLIPTextModelWithProjection:
['text_model.embeddings.position_ids']
The config attributes {'interpolation_type': 'linear', 'skip_prk_steps': True} were passed to LCMScheduler, but are not expected and will be ignored. Please verify your scheduler_config.json configuration file.
C:\Users\leafi\miniconda3\envs\comfyui\lib\site-packages\insightface\utils\transform.py:68: FutureWarning: rcond parameter will change to the default of machine precision times max(M, N) where M and N are the input matrix dimensions.
To use the future default and silence this warning we advise to pass rcond=None, to keep using the old, explicitly pass rcond=-1.
P = np.linalg.lstsq(X_homo, Y)[0].T # Affine matrix. 3 x 4
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:51<00:00, 25.80s/it]
C:\Users\leafi\miniconda3\envs\comfyui\lib\site-packages\diffusers\image_processor.py:92: RuntimeWarning: invalid value encountered in cast
images = (images * 255).round().astype("uint8")
Prompt executed in 274.22 seconds

[nosiu]

The first run should be slow, it's loading model, controlnet, LCM lora ... it takes time.
After that, you can change the mask, input image, and parameters in the generator node and it will be faster.
If you change the model, or disconnect LORA it will load it again (and it will be slow).