city96 / ComfyUI-GGUF

GGUF Quantization support for native ComfyUI models
Apache License 2.0
1.02k stars 67 forks source link

comfyui-gguf fails to run on macOS Sequoia while it woked fine on macOS sonoma. #108

Closed craii closed 1 month ago

craii commented 1 month ago

ComfyUI-GGUF works fine on macOS sonoma. But after I upgrade mac to Sequoia, it fails.

Reinstall comfyui-gguf, reinstall torch / numpy etc... all attemps fails.

here is the log

/Users/eliascheung/Documents/ComfyUI/custom_nodes/ComfyUI-GGUF/nodes.py:79: UserWarning: The given NumPy array is not writable, and PyTorch does not support non-writable tensors. This means writing to this tensor will result in undefined behavior. You may want to copy the array to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /Users/runner/work/pytorch/pytorch/pytorch/torch/csrc/utils/tensor_numpy.cpp:212.) torch_tensor = torch.from_numpy(tensor.data) # mmap

ggml_sd_loader: 1 472 2 304 model weight dtype torch.bfloat16, manual cast: None model_type FLOW Requested to load Flux Loading 1 new model loaded completely 0.0 6456.9610595703125 True 0%| | 0/4 [00:00<?, ?it/s]/AppleInternal/Library/BuildRoots/5a8a3fcc-55cb-11ef-848e-8a553ba56670/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:891: failed assertion `[MPSNDArray, initWithBufferImpl:offset:descriptor:isForNDArrayAlias:isUserBuffer:] Error: buffer is not large enough. Must be 63700992 bytes ' /Users/eliascheung/anaconda3/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown warnings.warn('resource_tracker: There appear to be %d '

city96 commented 1 month ago

Same issue as https://github.com/city96/ComfyUI-GGUF/issues/107 so tracking there, don't have any apple hardware so unlikely I'll be able to fix it but there's a pytorch issue/PR linked there that might.