PI-FLUX.2 Dev NVFP4 32B fails to load with state_dict size mismatch (FLUX.2 Dev 32B works)

What happened?
The standard FLUX.2 Dev 32B image model loads and generates images correctly (although it uses too much system RAM on my machine). When I switch the image model to PI-FLUX.2 Dev NVFP4 32B, the image prompts are generated successfully, but the image model never loads. Image generation does not start, and the terminal reports repeated PyTorch state_dict size mismatch errors.

Switching back to FLUX.2 Dev 32B immediately works again using the exact same project and settings. This appears to be specific to PI-FLUX.2 Dev NVFP4 32B, not the prompts or project itself.

Steps to reproduce

1. Launch Maestro.
2. Select PI-FLUX.2 Dev NVFP4 32B as the image model.
3. Generate a project (or regenerate image prompts). Use Director mode, but uncheck "Auto". Press generate when prompts appear ready.
4. Image prompts complete successfully.
5. When Maestro begins loading the image model, it fails with repeated state_dict size mismatch errors.
6. Switch back to FLUX.2 Dev 32B using the same project.
7. The model loads and image generation works normally.

Your system (OS / GPU / RAM / VRAM / etc.)
OS:* Windows 11

• GPU: NVIDIA GeForce RTX 5090
• VRAM: 32 GB
• System RAM: 64 GB
• Page File: 64 GB initial / 128 GB maximum
• Image model (working): FLUX.2 Dev 32B
• Image model (failing): PI-FLUX.2 Dev NVFP4 32B

Logs / full error output

Paste the full log output here
```[ERROR] Error(s) in loading state_dict for Flux:
size mismatch for double_blocks.0.img_mlp.0.weight:
copying a param with shape torch.Size([36864, 3072]) from checkpoint,
the shape in current model is torch.Size([36864, 6144]).

size mismatch for double_blocks.0.img_mlp.2.weight:
copying a param with shape torch.Size([6144, 9216]) from checkpoint,
the shape in current model is torch.Size([6144, 18432]).

...

The same state_dict size mismatch pattern repeats throughout many layers.