how to use quantization and device_map=balance to run qwen-image on kaggle T4 * 2

```python3
!python3 -m pip install -U diffusers peft bitsandbytes protobuf
import diffusers, torch, math
qwen = diffusers.QwenImagePipeline.from_pretrained('Qwen/Qwen-Image', quantization_config=diffusers.PipelineQuantizationConfig(quant_backend='bitsandbytes_4bit', quant_kwargs={'load_in_4bit':True, 'bnb_4bit_quant_type':'nf4', 'bnb_4bit_compute_dtype':torch.float16}, components_to_quantize=['transformer', 'text_encoder']), torch_dtype=torch.float16, device_map='balanced')
print(qwen.hf_device_map)
qwen.scheduler = diffusers.FlowMatchEulerDiscreteScheduler.from_config({'base_image_seq_len':256, 'base_shift':math.log(3), 'invert_sigmas':False, 'max_image_seq_len':8192, 'max_shift':math.log(3), 'num_train_timesteps':1000, 'shift':1, 'shift_terminal':None, 'stochastic_sampling':False, 'time_shift_type':'exponential', 'use_beta_sigmas':False, 'use_dynamic_shifting':True, 'use_exponential_sigmas':False, 'use_karras_sigmas':False})
qwen.load_lora_weights('lightx2v/Qwen-Image-Lightning', weight_name='Qwen-Image-Lightning-4steps-V2.0.safetensors', adapter_name='lightning')
qwen.set_adapters('lightning', adapter_weights=1)
qwen(prompt='a beautiful girl', height=1280, width=720, num_inference_steps=4, true_cfg_scale=1).images[0].save('a.png')
```

WARNING:accelerate.big_modeling:Some parameters are on the meta device because they were offloaded to the cpu.
{'text_encoder': 'cpu', 'vae': 0} where is the transformer ?

NotImplementedError: Cannot copy out of meta tensor; no data!

I want to ask how to make the above code work in kaggle. why 16G * 2 vram still not enough to run q4 quantization qwen-image? I want to take full advantage of 2 gpu. Do I need max_memory?

full error logs:
/usr/local/lib/python3.11/dist-packages/torch/utils/_contextlib.py in decorate_context(*args, **kwargs)
114 def decorate_context(*args, **kwargs):
115 with ctx_factory():
--> 116 return func(*args, **kwargs)
117
118 return decorate_context

/usr/local/lib/python3.11/dist-packages/diffusers/pipelines/qwenimage/pipeline_qwenimage.py in __call__(self, prompt, negative_prompt, true_cfg_scale, height, width, num_inference_steps, sigmas, guidance_scale, num_images_per_prompt, generator, latents, prompt_embeds, prompt_embeds_mask, negative_prompt_embeds, negative_prompt_embeds_mask, output_type, return_dict, attention_kwargs, callback_on_step_end, callback_on_step_end_tensor_inputs, max_sequence_length)
566 )
567 do_true_cfg = true_cfg_scale > 1 and has_neg_prompt
--> 568 prompt_embeds, prompt_embeds_mask = self.encode_prompt(
569 prompt=prompt,
570 prompt_embeds=prompt_embeds,

/usr/local/lib/python3.11/dist-packages/diffusers/pipelines/qwenimage/pipeline_qwenimage.py in encode_prompt(self, prompt, device, num_images_per_prompt, prompt_embeds, prompt_embeds_mask, max_sequence_length)
252
253 if prompt_embeds is None:
--> 254 prompt_embeds, prompt_embeds_mask = self._get_qwen_prompt_embeds(prompt, device)
255
256 prompt_embeds = prompt_embeds[:, :max_sequence_length]

/usr/local/lib/python3.11/dist-packages/diffusers/pipelines/qwenimage/pipeline_qwenimage.py in _get_qwen_prompt_embeds(self, prompt, device, dtype)
203 txt, max_length=self.tokenizer_max_length + drop_idx, padding=True, truncation=True, return_tensors="pt"
204 ).to(device)
--> 205 encoder_hidden_states = self.text_encoder(
206 input_ids=txt_tokens.input_ids,
207 attention_mask=txt_tokens.attention_mask,

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _wrapped_call_impl(self, *args, **kwargs)
1737 return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
1738 else:
-> 1739 return self._call_impl(*args, **kwargs)
1740
1741 # torchrec tests the code consistency with the following code

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1748 or _global_backward_pre_hooks or _global_backward_hooks
1749 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1750 return forward_call(*args, **kwargs)
1751
1752 result = None

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in new_forward(module, *args, **kwargs)
173 output = module._old_forward(*args, **kwargs)
174 else:
--> 175 output = module._old_forward(*args, **kwargs)
176 return module._hf_hook.post_forward(module, output)
177

/usr/local/lib/python3.11/dist-packages/transformers/utils/generic.py in wrapper(self, *args, **kwargs)
941
942 try:
--> 943 output = func(self, *args, **kwargs)
944 if is_requested_to_return_tuple or (is_configured_to_return_tuple and is_top_level_module):
945 output = output.to_tuple()

/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py in forward(self, input_ids, attention_mask, position_ids, past_key_values, inputs_embeds, labels, use_cache, output_attentions, output_hidden_states, pixel_values, pixel_values_videos, image_grid_thw, video_grid_thw, rope_deltas, cache_position, second_per_grid_ts, **kwargs)
1507 )
1508
-> 1509 outputs = self.model(
1510 input_ids=input_ids,
1511 pixel_values=pixel_values,

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _wrapped_call_impl(self, *args, **kwargs)
1737 return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
1738 else:
-> 1739 return self._call_impl(*args, **kwargs)
1740
1741 # torchrec tests the code consistency with the following code

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1748 or _global_backward_pre_hooks or _global_backward_hooks
1749 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1750 return forward_call(*args, **kwargs)
1751
1752 result = None

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in new_forward(module, *args, **kwargs)
173 output = module._old_forward(*args, **kwargs)
174 else:
--> 175 output = module._old_forward(*args, **kwargs)
176 return module._hf_hook.post_forward(module, output)
177

/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py in forward(self, input_ids, attention_mask, position_ids, past_key_values, inputs_embeds, use_cache, output_attentions, output_hidden_states, return_dict, pixel_values, pixel_values_videos, image_grid_thw, video_grid_thw, rope_deltas, cache_position, second_per_grid_ts, **kwargs)
1328 position_ids = position_ids.unsqueeze(0).expand(3, -1, -1)
1329
-> 1330 outputs = self.language_model(
1331 input_ids=None,
1332 position_ids=position_ids,

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _wrapped_call_impl(self, *args, **kwargs)
1737 return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
1738 else:
-> 1739 return self._call_impl(*args, **kwargs)
1740
1741 # torchrec tests the code consistency with the following code

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1748 or _global_backward_pre_hooks or _global_backward_hooks
1749 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1750 return forward_call(*args, **kwargs)
1751
1752 result = None

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in new_forward(module, *args, **kwargs)
173 output = module._old_forward(*args, **kwargs)
174 else:
--> 175 output = module._old_forward(*args, **kwargs)
176 return module._hf_hook.post_forward(module, output)
177

/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py in forward(self, input_ids, attention_mask, position_ids, past_key_values, inputs_embeds, use_cache, output_attentions, output_hidden_states, return_dict, cache_position, **kwargs)
918 all_hidden_states += (hidden_states,)
919
--> 920 layer_outputs = decoder_layer(
921 hidden_states,
922 attention_mask=causal_mask_mapping[decoder_layer.attention_type],

/usr/local/lib/python3.11/dist-packages/transformers/modeling_layers.py in __call__(self, *args, **kwargs)
81
82 return self._gradient_checkpointing_func(partial(super().__call__, **kwargs), *args)
---> 83 return super().__call__(*args, **kwargs)

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _wrapped_call_impl(self, *args, **kwargs)
1737 return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
1738 else:
-> 1739 return self._call_impl(*args, **kwargs)
1740
1741 # torchrec tests the code consistency with the following code

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1748 or _global_backward_pre_hooks or _global_backward_hooks
1749 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1750 return forward_call(*args, **kwargs)
1751
1752 result = None

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in new_forward(module, *args, **kwargs)
173 output = module._old_forward(*args, **kwargs)
174 else:
--> 175 output = module._old_forward(*args, **kwargs)
176 return module._hf_hook.post_forward(module, output)
177

/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py in forward(self, hidden_states, attention_mask, position_ids, past_key_value, output_attentions, use_cache, cache_position, position_embeddings, **kwargs)
775
776 # Self Attention
--> 777 hidden_states, self_attn_weights, present_key_value = self.self_attn(
778 hidden_states=hidden_states,
779 attention_mask=attention_mask,

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _wrapped_call_impl(self, *args, **kwargs)
1737 return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
1738 else:
-> 1739 return self._call_impl(*args, **kwargs)
1740
1741 # torchrec tests the code consistency with the following code

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1748 or _global_backward_pre_hooks or _global_backward_hooks
1749 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1750 return forward_call(*args, **kwargs)
1751
1752 result = None

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in new_forward(module, *args, **kwargs)
173 output = module._old_forward(*args, **kwargs)
174 else:
--> 175 output = module._old_forward(*args, **kwargs)
176 return module._hf_hook.post_forward(module, output)
177

/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py in forward(self, hidden_states, attention_mask, position_ids, past_key_value, output_attentions, use_cache, cache_position, position_embeddings, **kwargs)
681 bsz, q_len, _ = hidden_states.size()
682
--> 683 query_states = self.q_proj(hidden_states)
684 key_states = self.k_proj(hidden_states)
685 value_states = self.v_proj(hidden_states)

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _wrapped_call_impl(self, *args, **kwargs)
1737 return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
1738 else:
-> 1739 return self._call_impl(*args, **kwargs)
1740
1741 # torchrec tests the code consistency with the following code

/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1748 or _global_backward_pre_hooks or _global_backward_hooks
1749 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1750 return forward_call(*args, **kwargs)
1751
1752 result = None

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in new_forward(module, *args, **kwargs)
168
169 def new_forward(module, *args, **kwargs):
--> 170 args, kwargs = module._hf_hook.pre_forward(module, *args, **kwargs)
171 if module._hf_hook.no_grad:
172 with torch.no_grad():

/usr/local/lib/python3.11/dist-packages/accelerate/hooks.py in pre_forward(self, module, *args, **kwargs)
358 self.tied_pointers_to_remove.add((value.data_ptr(), self.execution_device))
359
--> 360 set_module_tensor_to_device(
361 module,
362 name,

/usr/local/lib/python3.11/dist-packages/accelerate/utils/modeling.py in set_module_tensor_to_device(module, tensor_name, device, value, dtype, fp16_statistics, tied_params_map, non_blocking, clear_cache)
361 new_value.SCB = new_value.SCB.to("cpu")
362 else:
--> 363 new_value = param_cls(new_value, requires_grad=old_value.requires_grad, **kwargs).to(
364 device, non_blocking=non_blocking
365 )

/usr/local/lib/python3.11/dist-packages/bitsandbytes/nn/modules.py in to(self, *args, **kwargs)
338 else:
339 if self.quant_state is not None:
--> 340 self.quant_state.to(device)
341
342 new_param = Params4bit(

/usr/local/lib/python3.11/dist-packages/bitsandbytes/functional.py in to(self, device)
537 def to(self, device):
538 # make sure the quantization state is on the right device
--> 539 self.code = self.code.to(device)
540 self.absmax = self.absmax.to(device)
541 if self.nested:

NotImplementedError: Cannot copy out of meta tensor; no data!

@yiyixuxu @DN6

Reproduce the supplied Kaggle script and inspect diffusers/pipelines/qwenimage/pipeline_qwenimage.py around encode_prompt, together with the printed qwen.hf_device_map and Accelerate warning. Trace the failure through the Qwen2.5-VL stack shown in the traceback; done means establishing whether the configuration can run across both GPUs and documenting the required behavior or limitation.

Written by the indexing model from the issue text.