Comfyui uses too much RAM after update

### Expected Behavior

Before update everything seems to work with sdxl and sd 1.5. I utilisied only default nodes

### Actual Behavior

Now default chekpoint loader node outputs an error when sdxl load requested

### Steps to Reproduce

The output error looks like this:

### Debug Logs

```powershell
Starting server

To see the GUI go to: http://127.0.0.1:8188
FETCH DATA from: D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\extension-node-map.json [DONE]
[]
[]
got prompt
!!! Exception during processing !!! Allocation on device
Traceback (most recent call last):
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 324, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 199, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 170, in _map_node_over_list
process_inputs(input_dict, i)
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 159, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\nodes.py", line 553, in load_checkpoint
out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 712, in load_checkpoint_guess_config
out = load_state_dict_guess_config(sd, output_vae, output_clip, output_clipvision, embedding_directory, output_model, model_options, te_model_options=te_model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 752, in load_state_dict_guess_config
model = model_config.get_model(sd, diffusion_model_prefix, device=inital_load_device)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\supported_models.py", line 208, in get_model
out = model_base.SDXL(self, model_type=self.model_type(state_dict, prefix), device=device)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\model_base.py", line 380, in __init__
super().__init__(model_config, model_type, device=device)
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\model_base.py", line 110, in __init__
self.diffusion_model = unet_model(**unet_config, device=device, operations=operations)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 788, in __init__
get_attention_layer(
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 568, in get_attention_layer
return SpatialTransformer(
^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 682, in __init__
[BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim[d],
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 682, in
[BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim[d],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 515, in __init__
self.ff = FeedForward(inner_dim, dim_out=dim, dropout=dropout, glu=gated_ff, dtype=dtype, device=device, operations=operations)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 80, in __init__
operations.Linear(inner_dim, dim_out, dtype=dtype, device=device)
File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\linear.py", line 106, in __init__
torch.empty((out_features, in_features), **factory_kwargs)
torch.OutOfMemoryError: Allocation on device

Got an OOM, unloading all loaded models.
Prompt executed in 0.43 seconds
```

### Other

This behaviour appeared only after update, previously everything worked properly. I have a 16 gb of RAM and RTX 4060 ti with 16 gb VRAM