search

fofr / cog-comfyui

Run ComfyUI with an API
https://replicate.com/fofr/any-comfyui-workflow
MIT License
468 stars 106 forks source link

majicmixRealistic_v7.safetensors model error #158

Open AfterRain1 opened 1 week ago

AfterRain1 commented 1 week ago

After updating today, I cannot use this model. An error message appears saying that the majicmixRealistic_v7.safetensors model cannot be referenced.It has always been available。

fofr commented 1 week ago

Is this still happening?

AfterRain1 commented 1 week ago

Yesterday it was majicmixRealistic_v7.safetensors, today the model reading failure was albedobaseXL_v21.safetensors,

I don't understand why such a problem occurs when running the same workflow in the same deployment. My GPU model is: Nvidia A40 (Large) GPU.

I ran this workflow twice, and the log analysis of the first successful run was as follows

... Executing node 602, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 ... Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple ... Executing node 104, title: Efficient Loader, class type: Efficient Loader ... Executing node 165, title: EasyLoraStack, class type: easy loraStack Executing node 2, title: Efficient Loader, class type: Efficient Loader ... Prompt executed in 41.07 seconds outputs: {'81': {'openpose_json': ...}, '108': {'images': ...}, '152': {'images': ...}, '4': {'images': ...}}

An error occurred during the second run and the image could not be output. The log analysis is as follows:

... Executing node 601, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 ... Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple !!! Exception during processing !!! Error while des erializing header: MetadataIncompleteBuffer Traceback (most recent call last): File "/src/ComfyUI/execution.py", line 313, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 188, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 165, in map_node_over_list process_inputs(input _dict, i) File "/src/ComfyUI/execution.py", line 154, in process_inputs results.append(getattr(obj, func)(**inputs)) File "/src/ComfyUI/nodes.py", line 539, in load_checkpoint out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings")) File "/src/ComfyUI/comfy/sd.py", line 510, in load_checkpoint_guess_config sd = comfy.utils.load_torch_file(ckpt_path) File "/src/ComfyUI/comfy/utils.py", line 34 , in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) File "/src/venv/lib/python3.10/site-packages/safetensors/torch.py", line 313, in load_file with safe_open(filename, framework="pt", device=device) as f: safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer Prompt executed in 0.91 seconds outputs: {}

fofr @.***> 于2024年9月3日周二 19:34写道：

Is this still happening?

— Reply to this email directly, view it on GitHub https://github.com/fofr/cog-comfyui/issues/158#issuecomment-2326292746, or unsubscribe https://github.com/notifications/unsubscribe-auth/A76TIWJVK3N4BTKB5IL3IADZUWNDJAVCNFSM6AAAAABNQNP4UKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRWGI4TENZUGY . You are receiving this because you authored the thread.Message ID: @.***>

AfterRain1 commented 1 week ago

Hi,there

https://huggingface.co/spaces/Kwai-Kolors/Kolors-Virtual-Try-On

Can this effect be increased in repllicate?

Thanks

dong mai @.***> 于2024年9月4日周三 13:18写道：

Yesterday it was majicmixRealistic_v7.safetensors, today the model reading failure was albedobaseXL_v21.safetensors,

I don't understand why such a problem occurs when running the same workflow in the same deployment. My GPU model is: Nvidia A40 (Large) GPU.

I ran this workflow twice, and the log analysis of the first successful run was as follows

... Executing node 602, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 ... Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple ... Executing node 104, title: Efficient Loader, class type: Efficient Loader ... Executing node 165, title: EasyLoraStack, class type: easy loraStack Executing node 2, title: Efficient Loader, class type: Efficient Loader ... Prompt executed in 41.07 seconds outputs: {'81': {'openpose_json': ...}, '108': {'images': ...}, '152': {'images': ...}, '4': {'images': ...}}

An error occurred during the second run and the image could not be output. The log analysis is as follows:

... Executing node 601, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 ... Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple !!! Exception during processing !!! Error while des erializing header: MetadataIncompleteBuffer Traceback (most recent call last): File "/src/ComfyUI/execution.py", line 313, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 188, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 165, in map_node_over_list process_inputs(input _dict, i) File "/src/ComfyUI/execution.py", line 154, in process_inputs results.append(getattr(obj, func)(**inputs)) File "/src/ComfyUI/nodes.py", line 539, in load_checkpoint out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings")) File "/src/ComfyUI/comfy/sd.py", line 510, in load_checkpoint_guess_config sd = comfy.utils.load_torch_file(ckpt_path) File "/src/ComfyUI/comfy/utils.py", line 34 , in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) File "/src/venv/lib/python3.10/site-packages/safetensors/torch.py", line 313, in load_file with safe_open(filename, framework="pt", device=device) as f: safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer Prompt executed in 0.91 seconds outputs: {}

fofr @.***> 于2024年9月3日周二 19:34写道：

Is this still happening?

— Reply to this email directly, view it on GitHub https://github.com/fofr/cog-comfyui/issues/158#issuecomment-2326292746, or unsubscribe https://github.com/notifications/unsubscribe-auth/A76TIWJVK3N4BTKB5IL3IADZUWNDJAVCNFSM6AAAAABNQNP4UKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRWGI4TENZUGY . You are receiving this because you authored the thread.Message ID: @.***>

AfterRain1 commented 1 week ago

Hi, I found the error logs I encountered twice, I hope it will help you understand the problem I encountered.

  1. Example： Predictions https://replicate.com/predictionsIDS:

a9asnc0yw5rgj0chpw8bb9qyfg

dong mai @.***> 于2024年9月6日周五 14:27写道：

Hi,there

https://huggingface.co/spaces/Kwai-Kolors/Kolors-Virtual-Try-On

Can this effect be increased in repllicate?

Thanks

dong mai @.***> 于2024年9月4日周三 13:18写道：

Yesterday it was majicmixRealistic_v7.safetensors, today the model reading failure was albedobaseXL_v21.safetensors,

I don't understand why such a problem occurs when running the same workflow in the same deployment. My GPU model is: Nvidia A40 (Large) GPU.

I ran this workflow twice, and the log analysis of the first successful run was as follows

... Executing node 602, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 ... Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple ... Executing node 104, title: Efficient Loader, class type: Efficient Loader ... Executing node 165, title: EasyLoraStack, class type: easy loraStack Executing node 2, title: Efficient Loader, class type: Efficient Loader ... Prompt executed in 41.07 seconds outputs: {'81': {'openpose_json': ...}, '108': {'images': ...}, '152': {'images': ...}, '4': {'images': ...}}

An error occurred during the second run and the image could not be output. The log analysis is as follows:

... Executing node 601, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 ... Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple !!! Exception during processing !!! Error while des erializing header: MetadataIncompleteBuffer Traceback (most recent call last): File "/src/ComfyUI/execution.py", line 313, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 188, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 165, in map_node_over_list process_inputs(input _dict, i) File "/src/ComfyUI/execution.py", line 154, in process_inputs results.append(getattr(obj, func)(**inputs)) File "/src/ComfyUI/nodes.py", line 539, in load_checkpoint out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings")) File "/src/ComfyUI/comfy/sd.py", line 510, in load_checkpoint_guess_config sd = comfy.utils.load_torch_file(ckpt_path) File "/src/ComfyUI/comfy/utils.py", line 34 , in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) File "/src/venv/lib/python3.10/site-packages/safetensors/torch.py", line 313, in load_file with safe_open(filename, framework="pt", device=device) as f: safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer Prompt executed in 0.91 seconds outputs: {}

fofr @.***> 于2024年9月3日周二 19:34写道：

Is this still happening?

— Reply to this email directly, view it on GitHub https://github.com/fofr/cog-comfyui/issues/158#issuecomment-2326292746, or unsubscribe https://github.com/notifications/unsubscribe-auth/A76TIWJVK3N4BTKB5IL3IADZUWNDJAVCNFSM6AAAAABNQNP4UKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRWGI4TENZUGY . You are receiving this because you authored the thread.Message ID: @.***>

第一次

Inputs uploaded to /tmp/inputs: input1.jpg Checking inputs ✅ /tmp/inputs/input1.jpg Checking weights Including weights for IPAdapter preset: FACEID PLUS V2 ✅ hand_pose_model.pth exists in ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators ✅ CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors exists in ComfyUI/models/clip_vision ✅ albedobaseXL_v21.safetensors exists in ComfyUI/models/checkpoints ✅ models/antelopev2 exists in ComfyUI/models/insightface ✅ 运动员 SD1.5.safetensors exists in ComfyUI/models/loras ✅ ip-adapter-faceid-plusv2_sd15_lora.safetensors exists in ComfyUI/models/loras ✅ ip-adapter-faceid-plusv2_sdxl_lora.safetensors exists in ComfyUI/models/loras ✅ ng_deepnegative_v1_75t.pt exists in ComfyUI/models/embeddings ✅ instantid-ip-adapter.bin exists in ComfyUI/models/instantid ✅ control_v11p_sd15_openpose.pth exists in ComfyUI/models/controlnet ✅ instantid-controlnet.safetensors exists in ComfyUI/models/controlnet ✅ facenet.pth exists in ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators ✅ ip-adapter-faceid-plusv2_sd15.bin exists in ComfyUI/models/ipadapter ✅ body_pose_model.pth exists in ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators ✅ ip-adapter-faceid-plusv2_sdxl.bin exists in ComfyUI/models/ipadapter ✅ majicmixRealistic_v7.safetensors exists in ComfyUI/models/checkpoints ✅ models/buffalo_l exists in ComfyUI/models/insightface Randomising seed to 229871032 Running workflow got prompt Executing node 601, title: Seed (rgthree), class type: Seed (rgthree) Executing node 397, title: Efficient Loader, class type: Efficient Loader !!! Exception during processing !!! Error while deserializing header: MetadataIncompleteBuffer Traceback (most recent call last): File "/src/ComfyUI/execution.py", line 313, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 188, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 165, in map_node_over_list process_inputs(input_dict, i) File "/src/ComfyUI/execution.py", line 154, in process_inputs results.append(getattr(obj, func)(**inputs)) File "/src/ComfyUI/custom_nodes/efficiency-nodes-comfyui/efficiency_nodes.py", line 158, in efficientloader model, clip = load_lora(lora_params, ckpt_name, my_unique_id, cache=lora_cache, ckpt_cache=ckpt_cache, cache_overwrite=True) File "/src/ComfyUI/custom_nodes/efficiency-nodes-comfyui/tsc_utils.py", line 370, in loadlora ckpt, clip, = load_checkpoint(ckpt_name, id, cache=ckpt_cache) File "/src/ComfyUI/custom_nodes/efficiency-nodes-comfyui/tsc_utils.py", line 246, in load_checkpoint out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings")) File "/src/ComfyUI/comfy/sd.py", line 510, in load_checkpoint_guess_config sd = comfy.utils.load_torch_file(ckpt_path) File "/src/ComfyUI/comfy/utils.py", line 34, in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) File "/src/venv/lib/python3.10/site-packages/safetensors/torch.py", line 313, in load_file with safe_open(filename, framework="pt", device=device) as f: safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer Prompt executed in 0.03 seconds outputs: {}

第二次

Inputs uploaded to /tmp/inputs: input1.jpg Checking inputs ✅ /tmp/inputs/input1.jpg Checking weights Including weights for IPAdapter preset: FACEID PLUS V2 ⏳ Downloading ip-adapter-faceid-plusv2_sdxl_lora.safetensors to ComfyUI/models/loras ✅ ip-adapter-faceid-plusv2_sdxl_lora.safetensors downloaded to ComfyUI/models/loras in 10.32s, size: 354.62MB ⏳ Downloading ip-adapter-faceid-plusv2_sd15_lora.safetensors to ComfyUI/models/loras ✅ ip-adapter-faceid-plusv2_sd15_lora.safetensors downloaded to ComfyUI/models/loras in 1.38s, size: 48.69MB ⏳ Downloading ng_deepnegative_v1_75t.pt to ComfyUI/models/embeddings ✅ ng_deepnegative_v1_75t.pt downloaded to ComfyUI/models/embeddings in 0.13s, size: 0.22MB ⏳ Downloading control_v11p_sd15_openpose.pth to ComfyUI/models/controlnet ✅ control_v11p_sd15_openpose.pth downloaded to ComfyUI/models/controlnet in 397.11s, size: 1378.28MB ⏳ Downloading hand_pose_model.pth to ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators ✅ hand_pose_model.pth downloaded to ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators in 0.40s, size: 140.52MB ⏳ Downloading body_pose_model.pth to ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators ✅ body_pose_model.pth downloaded to ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators in 0.51s, size: 199.57MB ✅ CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors exists in ComfyUI/models/clip_vision ✅ instantid-controlnet.safetensors exists in ComfyUI/models/controlnet ⏳ Downloading ip-adapter-faceid-plusv2_sdxl.bin to ComfyUI/models/ipadapter ✅ ip-adapter-faceid-plusv2_sdxl.bin downloaded to ComfyUI/models/ipadapter in 55.07s, size: 1418.64MB ✅ 运动员 SD1.5.safetensors exists in ComfyUI/models/loras ✅ majicmixRealistic_v7.safetensors exists in ComfyUI/models/checkpoints ✅ models/buffalo_l exists in ComfyUI/models/insightface ✅ albedobaseXL_v21.safetensors exists in ComfyUI/models/checkpoints ⏳ Downloading facenet.pth to ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators ✅ facenet.pth downloaded to ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts/lllyasviel/Annotators in 0.39s, size: 146.60MB ✅ instantid-ip-adapter.bin exists in ComfyUI/models/instantid ✅ models/antelopev2 exists in ComfyUI/models/insightface ⏳ Downloading ip-adapter-faceid-plusv2_sd15.bin to ComfyUI/models/ipadapter ✅ ip-adapter-faceid-plusv2_sd15.bin downloaded to ComfyUI/models/ipadapter in 0.39s, size: 149.31MB Randomising seed to 1747561316 Running workflow got prompt Executing node 601, title: Seed (rgthree), class type: Seed (rgthree) Executing node 444, title: Load Image, class type: LoadImage Executing node 568, title: Image Resize by Longer Side, class type: JWImageResizeByLongerSide Executing node 569, title: LayerMask: PersonMaskUltra V2, class type: LayerMask: PersonMaskUltra V2 I0000 00:00:1725413752.105590 861 gl_context_egl.cc:85] Successfully initialized EGL. Major : 1 Minor: 5 I0000 00:00:1725413752.123586 13373 gl_context.cc:357] GL version: 3.2 (OpenGL ES 3.2 NVIDIA 535.177), renderer: NVIDIA A40/PCIe/SSE2 W0000 00:00:1725413752.145218 13376 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.😺dzNodes: LayerStyle -> [1;33mvitmatte processing, image size = 683x1024, device = cuda.[m 😺dzNodes: LayerStyle -> [1;32mPersonMaskUltra V2 Processed 1 image(s).[m Executing node 572, title: Bounded Image Crop with Mask, class type: Bounded Image Crop with Mask Executing node 567, title: Load Checkpoint, class type: CheckpointLoaderSimple !!! Exception during processing !!! Error while deserializing header: MetadataIncompleteBuffer Traceback (most recent call last): File "/src/ComfyUI/execution.py", line 313, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 188, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "/src/ComfyUI/execution.py", line 165, in map_node_over_list process_inputs(input_dict, i) File "/src/ComfyUI/execution.py", line 154, in process_inputs results.append(getattr(obj, func)(**inputs)) File "/src/ComfyUI/nodes.py", line 539, in load_checkpoint out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings")) File "/src/ComfyUI/comfy/sd.py", line 510, in load_checkpoint_guess_config sd = comfy.utils.load_torch_file(ckpt_path) File "/src/ComfyUI/comfy/utils.py", line 34, in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) File "/src/venv/lib/python3.10/site-packages/safetensors/torch.py", line 313, in load_file with safe_open(filename, framework="pt", device=device) as f: safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer Prompt executed in 0.91 seconds outputs: {}