runtime error

Exit code: 1. Reason: 1.09G [00:03<00:00, 24.1MB/s]  Download complete: 100%|██████████| 1.09G/1.09G [00:03<00:00, 304MB/s] saved as: /root/.cache/huggingface/hub/models--Qwen--Qwen3-30B-A3B-Instruct-2507/snapshots/0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe/splitted_model/model.layers.47.safetensors saved as: /root/.cache/huggingface/hub/models--Qwen--Qwen3-30B-A3B-Instruct-2507/snapshots/0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe/splitted_model/model.norm.safetensors saved as: /root/.cache/huggingface/hub/models--Qwen--Qwen3-30B-A3B-Instruct-2507/snapshots/0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe/splitted_model/lm_head.safetensors 96%|█████████▌| 49/51 [03:56<00:07, 3.86s/it] 100%|██████████| 51/51 [03:56<00:00, 4.64s/it] [transformers] `torch_dtype` is deprecated! Use `dtype` instead! built empty Qwen3MoeForCausalLM via AutoModelForCausalLM (attn=sdpa) Traceback (most recent call last): File "/app/app.py", line 6, in <module> model = AutoModel.from_pretrained(MODEL_NAME) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/site-packages/airllm/auto_model.py", line 63, in from_pretrained return class_(pretrained_model_name_or_path, *inputs, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/site-packages/airllm/airllm_base.py", line 194, in __init__ self.init_model() File "/usr/local/lib/python3.12/site-packages/airllm/airllm_base.py", line 372, in init_model set_module_tensor_to_device(self.model, buffer_name, self.running_device, value=buffer) File "/usr/local/lib/python3.12/site-packages/accelerate/utils/modeling.py", line 353, in set_module_tensor_to_device new_value = value.to(device, non_blocking=non_blocking) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/site-packages/torch/cuda/__init__.py", line 529, in _lazy_init torch._C._cuda_init() RuntimeError: No CUDA GPUs are available

Container logs: