Commit History

  1. Try _fast_init=False to prevent meta device initialization Running

    ACE-Step Custom
    5b76ce1

  1. Fix dtype parameter: use dtype instead of torch_dtype, remove device_map

    ACE-Step Custom
    140e1e2
  2. Fix meta tensor: add device_map=None and use torch_dtype parameter

    ACE-Step Custom
    a11c404
  3. Fix meta tensor error: disable low_cpu_mem_usage for model loading

    ACE-Step Custom
    23dc982
  4. Add vector-quantize-pytorch dependency for ACE-Step model

    ACE-Step Custom
    5e842f6
  5. Fix DiT initialization: pass model name only, not full path

    ACE-Step Custom
    6a590ee
  6. Fix TypeError: pass Path object to ensure_main_model, not string

    ACE-Step Custom
    fa7f63d
  7. Fix syntax error: close unterminated docstring in _load_models

    ACE-Step Custom
    78910e3
  8. Use official ACE-Step model downloader for automatic model downloads

    ACE-Step Custom
    ee19acb
  9. Monkey patch get_api_info to prevent JSON schema errors

    ACE-Step Custom
    55a1124
  10. Move endpoint override before launch

    ACE-Step Custom
    ccb3542
  11. Fix fastapi_app attribute name and move endpoint override after launch

    ACE-Step Custom
    2217362
  12. Add /info endpoint override and pin Gradio version

    ACE-Step Custom
    d6ad90b
  13. Add loguru dependency and fix Gradio 4.0.0 compatibility

    ACE-Step Custom
    9259263
  14. Fix engine initialization and audio I/O components

    ACE-Step Custom
    4709141
  15. Deploy ACE-Step Custom Edition with bug fixes

    ACE-Step Custom
    a602628