VAEDBCO0G MPS backend out of memory

### Custom Node Testing

- [x] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

I used the M3 MAX 128GB version and calculated that it would last approximately 8 hours. The GPU usage rate remained around 80%. Then it prompted me that the MPS memory was insufficient. Could you please tell me how to solve this problem?

### Actual Behavior

### Steps to Reproduce

[video_wan2_2_5B_ti2v.json](https://github.com/user-attachments/files/21745804/video_wan2_2_5B_ti2v.json)

### Debug Logs

```powershell
RuntimeError: MPS backend out of memory (MPS allocated: 22.57 GiB, other allocations: 139.74 GiB, max allowed: 163.20 GiB). Tried to allocate 1.30 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).

Prompt executed in 01:22:17
```

### Other

_No response_