Sending subsequent prompts containing the exact same configuration for UnifiedLoaderFacedID, it takes a long time for it to load all the models, each time though they should be still present:
2024-05-03T20:51:42.652083183Z got prompt
2024-05-03T20:51:43.210435765Z INFO: IPAdapter model loaded from /workspace/models/ipadapter/ip-adapter-faceid-plusv2_sdxl.bin
2024-05-03T20:51:43.458743811Z INFO: LoRA model loaded from /workspace/models/loras/ip-adapter-faceid-plusv2_sdxl_lora.safetensors
2024-05-03T20:51:43.736389380Z Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
2024-05-03T20:52:36.323684443Z find model: /workspace/ComfyUI/models/insightface/models/buffalo_l/1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0
2024-05-03T20:52:36.357869568Z Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
2024-05-03T20:52:38.197729795Z find model: /workspace/ComfyUI/models/insightface/models/buffalo_l/2d106det.onnx landmark_2d_106 ['None', 3, 192, 192] 0.0 1.0
2024-05-03T20:52:38.271701100Z Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
2024-05-03T20:52:38.271718430Z find model: /workspace/ComfyUI/models/insightface/models/buffalo_l/det_10g.onnx detection [1, 3, '?', '?'] 127.5 128.0
2024-05-03T20:52:38.293622070Z Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
2024-05-03T20:52:38.820498750Z find model: /workspace/ComfyUI/models/insightface/models/buffalo_l/genderage.onnx genderage ['None', 3, 96, 96] 0.0 1.0
2024-05-03T20:52:39.015671372Z Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
2024-05-03T20:52:39.129008964Z find model: /workspace/ComfyUI/models/insightface/models/buffalo_l/w600k_r50.onnx recognition ['None', 3, 112, 112] 127.5 127.5
2024-05-03T20:52:39.129026155Z set det-size: (640, 640)
2024-05-03T20:52:39.129032625Z INFO: InsightFace model loaded with CUDA provider
2024-05-03T20:52:41.596427807Z Requested to load CLIPVisionModelProjection
2024-05-03T20:52:41.596463188Z Loading 1 new model
2024-05-03T20:52:53.629798610Z Requested to load SDXL
2024-05-03T20:52:53.629828051Z Loading 1 new model
Also there is only the CPUExceutionProvider mentioned allthough I am running on gpu-only and onnxruntime-gpu is installed.
Sending subsequent prompts containing the exact same configuration for UnifiedLoaderFacedID, it takes a long time for it to load all the models, each time though they should be still present:
Also there is only the CPUExceutionProvider mentioned allthough I am running on
gpu-onlyandonnxruntime-gpuis installed.Plenty of VRAM is available.
ComfyUI and IPAdapter_plus are both up to date.