Help, I'm going crazy.

Every time I click "Queue Prompt" to submit, I have to wait for a long time.

The cmd shows:

`MemoryEfficientCrossAttention. Query dim is 1280, context_dim is None and using 20 heads.Setting up MemoryEfficientCrossAttention. Query dim is 1280, context_dim is 2048 and using 20 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 1280, context_dim is None and using 20 heads.Setting up MemoryEfficientCrossAttention. Query dim is 1280, context_dim is 2048 and using 20 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 1280, context_dim is None and using 20 heads.Setting up MemoryEfficientCrossAttention. Query dim is 1280, context_dim is 2048 and using 20 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is None and using 10 heads.Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is 2048 and using 10 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is None and using 10 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is 2048 and using 10 heads.Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is None and using 10 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is 2048 and using 10 heads.Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is None and using 10 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is 2048 and using 10 heads.Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is None and using 10 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is 2048 and using 10 heads.Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is None and using 10 heads.
Setting up MemoryEfficientCrossAttention. Query dim is 640, context_dim is 2048 and using 10 heads.model_type EPS
adm 2816making attention of type 'vanilla-xformers' with 512 in_channels
building MemoryEfficientAttnBlock with 512 in_channels...Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type 'vanilla-xformers' with 512 in_channelsbuilding MemoryEfficientAttnBlock with 512 in_channels...
left over keys: dict_keys(['cond_stage_model.clip_l.transformer.text_model.embeddings.position_ids'])loading new
loading newloading in lowvram mode 879.826865196228
100%|###########################################################################################| 24/24 [02:10<00:00, 5.44s/it]Prompt executed in 322.20 seconds`

And I have to wait for five or six minutes each time before it starts running. Later, I found that this only happens when switching checkpoints. If I stick to using the same checkpoint, it's only slow the first time, and then it becomes fast. But if I switch checkpoints in "Load Checkpoint," it will still be slow. How can I solve this issue? I have an Nvidia 2080 Super with 16GB RAM and an SSD hard drive.