Spaces:
Running on Zero
Running on Zero
Commit History
-
Force traditional loading: low_cpu_mem_usage=False + explicit device move Running
ACE-Step Custom180e887 -
Use official ACE-Step demo approach: dtype as string (bfloat16)
ACE-Step Custom897bbd6 -
Use device_map to force CUDA placement and prevent meta device
ACE-Step Custom2538e84
-
Fix: Force CPU device context during model init to prevent meta tensor operations
ACE-Step Customfc9be45 -
Fix: Remove device_map to prevent meta tensor initialization - ACE-Step custom model needs direct device placement
ACE-Step Customaa918f7
-
Fix: Add device_map to prevent meta tensor errors on ZeroGPU
ACE-Step Custom- Added explicit device_map parameter to all model loading calls - Fixes 'Tensor.item() cannot be called on meta tensors' error - Ensures models load directly to target device on HF Spaces - Applies to DiT, VAE, Text Encoder, and LLM models6b39c2d -
Try _fast_init=False to prevent meta device initialization
ACE-Step Custom5b76ce1
-
Fix dtype parameter: use dtype instead of torch_dtype, remove device_map
ACE-Step Custom140e1e2 -
Fix meta tensor: add device_map=None and use torch_dtype parameter
ACE-Step Customa11c404 -
Fix meta tensor error: disable low_cpu_mem_usage for model loading
ACE-Step Custom23dc982 -
Deploy ACE-Step Custom Edition with bug fixes
ACE-Step Customa602628