You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When I run my workflow comfy-workflow-api.json on Replicate using version d485be22dbc, I get an error:
Prediction failed.
Command '['pget', '--log-level', 'warn', '-xf', 'https://weights.replicate.delivery/default/comfy-ui/facedetection/detection_Resnet50_Final.pth.tar', '/srv/r8/monobase/monobase/g00012/python3.10-torch2.5.1-cu121/lib/python3.10/site-packages/facexlib/weights']' returned non-zero exit status 1.
Logs:
Error: error writing file: error extracting file: mkdir /srv/r8/monobase/monobase/g00012/python3.10-torch2.5.1-cu121/lib/python3.10/site-packages/facexlib: read-only file system
Traceback (most recent call last):
File "/srv/r8/monobase/cog/g1df235d5/cog0.13.6-python3.10/lib/python3.10/site-packages/cog/server/worker.py", line 646, in _handle_predict_error
yield
File "/srv/r8/monobase/cog/g1df235d5/cog0.13.6-python3.10/lib/python3.10/site-packages/cog/server/worker.py", line 538, in _predict
result = predict(**payload)
File "/src/predict.py", line 127, in predict
wf = self.comfyUI.load_workflow(workflow_json or EXAMPLE_WORKFLOW_JSON)
File "/src/comfyui.py", line 275, in load_workflow
self.handle_weights(wf)
File "/src/comfyui.py", line 98, in handle_weights
self.apply_helper_methods("add_weights", weights_to_download, Node(node))
File "/src/comfyui.py", line 86, in apply_helper_methods
method(*args, **kwargs)
File "/src/custom_node_helpers/PuLID.py", line 41, in add_weights
weights_downloader.download_if_not_exists(
File "/src/weights_downloader.py", line 61, in download_if_not_exists
WeightsDownloader.download(weight_str, url, dest)
File "/src/weights_downloader.py", line 72, in download
subprocess.check_call(
File "/srv/r8/monobase/uv/python/cpython-3.10.15-linux-x86_64-gnu/lib/python3.10/subprocess.py", line 369, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['pget', '--log-level', 'warn', '-xf', 'https://weights.replicate.delivery/default/comfy-ui/facedetection/detection_Resnet50_Final.pth.tar', '/srv/r8/monobase/monobase/g00012/python3.10-torch2.5.1-cu121/lib/python3.10/site-packages/facexlib/weights']' returned non-zero exit status 1.
====================================
Inputs uploaded to /tmp/inputs:
input.jpg
====================================
Checking inputs
✅ /tmp/inputs/input.jpg
====================================
Checking weights
Adding ltx-video-2b-v0.9.1.safetensors to CHECKPOINTS
Adding realvisxlV50_v50Bakedvae.safetensors to CHECKPOINTS
Adding CN-anytest_v3-50000_fp16.safetensors to CONTROLNET
Adding CogFlorence-2.1-Large to LLM
Adding CogFlorence-2.2-Large to LLM
Adding Florence-2-base to LLM
Adding Florence-2-base-ft to LLM
Adding Florence-2-base-PromptGen-v1.5 to LLM
Adding Florence-2-base-PromptGen-v2.0 to LLM
Adding Florence-2-DocVQA to LLM
Adding Florence-2-Flux-Large to LLM
Adding Florence-2-large to LLM
Adding Florence-2-large-ft to LLM
Adding Florence-2-large-PromptGen-v1.5 to LLM
Adding Florence-2-large-PromptGen-v2.0 to LLM
Adding Florence-2-SD3-Captioner to LLM
⏳ Downloading detection_Resnet50_Final.pth to /srv/r8/monobase/monobase/g00012/python3.10-torch2.5.1-cu121/lib/python3.10/site-packages/facexlib/weights
I guess that tjis dependency on detection_Resnet50_Final is added by one of the PULID nodes that I use with FLUX. If I remove PULID nodes from my workflow, it works. Actually, I don't understand the error. The file URL is accessible, I can download it manually.
The text was updated successfully, but these errors were encountered:
When I run my workflow comfy-workflow-api.json on Replicate using version
d485be22dbc
, I get an error:Logs:
I guess that tjis dependency on detection_Resnet50_Final is added by one of the PULID nodes that I use with FLUX. If I remove PULID nodes from my workflow, it works. Actually, I don't understand the error. The file URL is accessible, I can download it manually.
The text was updated successfully, but these errors were encountered: