[BUG] Qwen3: model loading failed when using meta device

Describe the bug
I am running on a single node with 4 GPUs; each GPU has 24GB GPU memory.

With Deepspeed-Inference, I was trying to load Qwen/Qwen3-4B using meta device. However, the loading failed and I got the following error:

NotImplementedError: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.

Although this small model doesn't need meta device, my ultimate goal is to use the bigger qwen3 models.

To Reproduce
Steps to reproduce the behavior:

1. Simple inference script to reproduce.

First of all, download Qwen/Qwen3-4B to local directory "Qwen3-4B"

Then, put the following code snippet to "qwen3_meta_device.py"

import os

import deepspeed
import torch
from transformers import AutoConfig, AutoModelForCausalLM

kwargs = {"torch_dtype": torch.float16}

model_config = AutoConfig.from_pretrained("./Qwen3-4B", **kwargs)

with deepspeed.OnDevice(dtype=kwargs["torch_dtype"], device="meta", enabled=True):
    model = AutoModelForCausalLM.from_config(model_config, **kwargs)

ds_inference_config = {
    "dtype": kwargs["torch_dtype"],
    "replace_with_kernel_inject": False,
    "tensor_parallel": {
        "tp_size": int(os.getenv("WORLD_SIZE", "1"))
    },
    "checkpoint": {
        "checkpoints": [
            "./Qwen3-4B/model-00001-of-00003.safetensors",
            "./Qwen3-4B/model-00002-of-00003.safetensors",
            "./Qwen3-4B/model-00003-of-00003.safetensors"
        ],
        "type": "DS_MODEL",
        "version": 1.0
    }
}

ds_engine = deepspeed.init_inference(model, config=ds_inference_config)
model = ds_engine.module
model.eval()

Finally, run "accelerate launch qwen3_meta_device.py"

2. What packages are required and their versions
   torch==2.5.1
   transformers==4.51.3
   deepspeed==0.16.7
   accelerate==1.6.0

3. How to run the script
   Put the above code snippet into this file: qwen3_meta_device.py
   Then, run the following:
   accelerate launch qwen3_meta_device.py

4. ...
   Expected behavior
   The model is expected to load successfully.

The code works fine for qwen2.5-7b-instruct (after replacing the checkpoint files in the config).

ds_report output
Please run ds_report to give us details about your setup.

Screenshots
If applicable, add screenshots to help explain your problem.

System info (please complete the following information):

• OS: [e.g. Ubuntu 18.04]
• GPU count and types [e.g. two machines with x8 A100s each]
• (if applicable) what DeepSpeed-MII version are you using
• (if applicable) Hugging Face Transformers/Accelerate/etc. versions
• Python version
• Any other relevant info about your setup

Docker context
Are you using a specific docker image that you can share?

Additional context
Add any other context about the problem here.

Start with qwen3_meta_device.py and the deepspeed.OnDevice and deepspeed.init_inference entry points used in the reproduction. Run the provided accelerate launch command with the listed package versions and Qwen3-4B checkpoints, then compare the meta-device loading path with the working Qwen2.5 example. Done means the model loads successfully without the reported meta-tensor error.