Qwen/Qwen2.5-1.5B-Instruct · [MPSTemporaryNDArray initWithDevice:descriptor:] Error: total bytes of NDArray

(transformers-test) ➜  transformers-test git:(main) ✗ uv run hello.py
config.json: 660B [00:00, 919kB/s]
model.safetensors: 100%|██████████████████████████████████████████████| 3.09G/3.09G [06:05<00:00, 8.46MB/s]
generation_config.json: 100%|██████████████████████████████████████████████| 242/242 [00:00<00:00, 835kB/s]
tokenizer_config.json: 7.30kB [00:00, 6.11MB/s]
vocab.json: 2.78MB [00:01, 1.93MB/s]
merges.txt: 1.67MB [00:00, 2.13MB/s]
tokenizer.json: 7.03MB [00:02, 2.70MB/s]
/AppleInternal/Library/BuildRoots/ce725a5f-c761-11ee-a4ec-b6ef2fd8d87b/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:788: failed assertion `[MPSTemporaryNDArray initWithDevice:descriptor:] Error: total bytes of NDArray > 2**32'
/Users/aaa/.local/share/uv/python/cpython-3.12.9-macos-aarch64-none/lib/python3.12/multiprocessing/resource_tracker.py:255: 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 '

i just run the demo code in my mac, is anyone know how to resove this?

requires-python = ">=3.12"
dependencies = [
    "accelerate>=1.7.0",
    "torch>=2.7.0",
    "transformers>=4.52.3",
]