runtime error

Exit code: 1. Reason: _image_edit_2509.sa(…): downloading bytes: ██████████| 245MB, 23.4MB/s apply_texture_v2_qwen_image_edit_2509.sa(…): reconstructing file: 100%|██████████| 295MB / 295MB, 28.7MB/s Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): reconstructing file: 0%| | 0.00B / 850MB  Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): downloading bytes: | 0.00B Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): downloading bytes: ████████ | 681MB, 50.0MB/s Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): reconstructing file: 39%|███▉ | 335MB / 850MB, 18.4kB/s  Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): downloading bytes: ██████████| 712MB, 68.0MB/s Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): downloading bytes: ██████████| 712MB, 68.0MB/s Qwen-Image-Lightning-4steps-V2.0-bf16.sa(…): reconstructing file: 100%|██████████| 850MB / 850MB, 82.0MB/s /usr/local/lib/python3.10/site-packages/peft/tuners/tuners_utils.py:305: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing! warnings.warn( Traceback (most recent call last): File "/home/user/app/app.py", line 53, in <module> pipe.transformer.set_attn_processor(QwenDoubleStreamAttnProcessorFA3()) File "/home/user/app/qwenimage/qwen_fa3_processor.py", line 57, in __init__ _ensure_fa3_available() File "/home/user/app/qwenimage/qwen_fa3_processor.py", line 20, in _ensure_fa3_available raise ImportError( ImportError: FlashAttention-3 via Hugging Face `kernels` is required. Tried `get_kernel('kernels-community/vllm-flash-attn3')` and failed with: A kernel version or revision must be specified. Use `version=<major>` for a stable kernel API version or `revision=<branch/tag/commit>` for an explicit Hub revision. See: https://huggingface.co/docs/kernels/migration

Container logs: