Does not produce a second image

### Custom Node Testing

- [ ] 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 expected an image to be made

[Workflow.txt](https://github.com/user-attachments/files/30132675/Workflow.txt)

### Actual Behavior

The first image was created. Second one was as shown in the attached file. Every file afterwards was a repetition of this broken image, no matter what template I used or what prompt I gave.

### Steps to Reproduce

Load the latina girl image template, run an image off. repeat. On repeat, I get that split screen weird coloured image.

### Debug Logs

```powershell
D:\Programs\AI Stuff\ComfyUI-amd>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --enable-dynamic-vram
[INFO] setup plugin alembic.autogenerate.schemas
[INFO] setup plugin alembic.autogenerate.tables
[INFO] setup plugin alembic.autogenerate.types
[INFO] setup plugin alembic.autogenerate.constraints
[INFO] setup plugin alembic.autogenerate.defaults
[INFO] setup plugin alembic.autogenerate.comments
[INFO] Found comfy_kitchen backend triton: {'available': False, 'disabled': True, 'unavailable_reason': "ImportError: No module named 'triton'", 'capabilities': []}
[INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'apply_rope_split_half', 'apply_rope_split_half1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'gemv_awq_w4a16', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8']}
[INFO] Found comfy_kitchen backend cuda: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'apply_rope_split_half', 'apply_rope_split_half1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'gemv_awq_w4a16', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8']}
[INFO] Checkpoint files will always be loaded safely.
[INFO] Total VRAM 34669 MB, total RAM 48761 MB
[INFO] pytorch version: 2.9.1+rocm7.2.1
[INFO] Set: torch.backends.cudnn.enabled = False for better AMD performance.
[INFO] AMD arch: gfx1150
[INFO] ROCm version: (7, 2)
[INFO] Set vram state to: NORMAL_VRAM
[INFO] Device: cuda:0 AMD Radeon(TM) 890M Graphics : native
[INFO] Using async weight offloading with 2 streams
[INFO] Enabled pinned memory 19504.0
[INFO] Using pytorch attention
aimdo: src-win/cuda-detour.c:38:INFO:aimdo_setup_hooks: installing 6 hooks
aimdo: src-win/cuda-detour.c:28:DEBUG:install_hook_entries: hooks successfully installed
aimdo: src-win/shmem-detect.c:80:INFO:comfy-aimdo WDDM adapter match: AMD Radeon(TM) 890M Graphics runtime_luid=00000000:0001042c dxgi_luid=00000000:0001042c
aimdo: src/control.c:248:INFO:comfy-aimdo inited for GPU: AMD Radeon(TM) 890M Graphics (VRAM: 34669 MB)
[INFO] DynamicVRAM support detected and enabled
[INFO] Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
[INFO] ComfyUI version: 0.26.0
[INFO] comfy-aimdo version: 0.4.10
[INFO] comfy-kitchen version: 0.2.10
[INFO] comfyui-frontend-package version: 1.45.19
[INFO] comfyui-workflow-templates version: 0.10.3
[INFO] comfyui-embedded-docs version: 0.5.5
[INFO] comfy-kitchen version: 0.2.10
[INFO] comfy-aimdo version: 0.4.10
[INFO] [Prompt Server] web root: D:\Programs\AI Stuff\ComfyUI-amd\python_embeded\Lib\site-packages\comfyui_frontend_package\static
[INFO] Asset seeder disabled
[INFO]
Import times for custom nodes:
[INFO] 0.0 seconds: D:\Programs\AI Stuff\ComfyUI-amd\ComfyUI\custom_nodes\websocket_image_save.py
[INFO]
[INFO] Context impl SQLiteImpl.
[INFO] Will assume non-transactional DDL.
[INFO] Using RAM pressure cache.
[INFO] Starting server

[INFO] To see the GUI go to: http://127.0.0.1:8188
[INFO] got prompt
[INFO] Using split attention in VAE
[INFO] Using split attention in VAE
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
[INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
[INFO] Requested to load ZImageTEModel_
[INFO] Model ZImageTEModel_ prepared for dynamic VRAM loading. 7671MB Staged. 0 patches attached. Force pre-loaded 145 weights: 383 KB.
D:\Programs\AI Stuff\ComfyUI-amd\ComfyUI\comfy\ops.py:60: UserWarning: Using AOTriton backend for Efficient Attention forward... (Triggered internally at C:/b/pytorch/aten/src/ATen/native/transformers/hip/attention.hip:1452.)
return torch.nn.functional.scaled_dot_product_attention(q, k, v, *args, **kwargs)
[INFO] model weight dtype torch.bfloat16, manual cast: None
[INFO] model_type FLOW
[INFO] Requested to load Lumina2
[INFO] Model Lumina2 prepared for dynamic VRAM loading. 11738MB Staged. 0 patches attached. Force pre-loaded 205 weights: 1045 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [01:31<00:00, 11.40s/it]
[INFO] Requested to load AutoencodingEngine
[INFO] 0 models unloaded.
[INFO] Model AutoencodingEngine prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. Force pre-loaded 108 weights: 182 KB.
[INFO] Prompt executed in 122.53 seconds
[INFO] got prompt
[INFO] Model ZImageTEModel_ prepared for dynamic VRAM loading. 7671MB Staged. 0 patches attached. Force pre-loaded 145 weights: 383 KB.
[INFO] Model Lumina2 prepared for dynamic VRAM loading. 11738MB Staged. 0 patches attached. Force pre-loaded 205 weights: 1045 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 9.76it/s]
[INFO] 0 models unloaded.
[INFO] Model AutoencodingEngine prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. Force pre-loaded 108 weights: 182 KB.
D:\Programs\AI Stuff\ComfyUI-amd\ComfyUI\nodes.py:1659: RuntimeWarning: invalid value encountered in cast
img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))
[INFO] Prompt executed in 11.55 seconds
[INFO] got prompt
[INFO] Model ZImageTEModel_ prepared for dynamic VRAM loading. 7671MB Staged. 0 patches attached. Force pre-loaded 145 weights: 383 KB.
[INFO] Model Lumina2 prepared for dynamic VRAM loading. 11738MB Staged. 0 patches attached. Force pre-loaded 205 weights: 1045 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 21.96it/s]
[INFO] 0 models unloaded.
[INFO] Model AutoencodingEngine prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. Force pre-loaded 108 weights: 182 KB.
[INFO] Prompt executed in 3.49 seconds
[INFO] got prompt
[INFO] Model ZImageTEModel_ prepared for dynamic VRAM loading. 7671MB Staged. 0 patches attached. Force pre-loaded 145 weights: 383 KB.
[INFO] Model Lumina2 prepared for dynamic VRAM loading. 11738MB Staged. 0 patches attached. Force pre-loaded 205 weights: 1045 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 22.35it/s]
[INFO] 0 models unloaded.
[INFO] Model AutoencodingEngine prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. Force pre-loaded 108 weights: 182 KB.
[INFO] Prompt executed in 3.10 seconds
[INFO] got prompt
[INFO] Model ZImageTEModel_ prepared for dynamic VRAM loading. 7671MB Staged. 0 patches attached. Force pre-loaded 145 weights: 383 KB.
[INFO] Model Lumina2 prepared for dynamic VRAM loading. 11738MB Staged. 0 patches attached. Force pre-loaded 205 weights: 1045 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 21.97it/s]
[INFO] 0 models unloaded.
[INFO] Model AutoencodingEngine prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. Force pre-loaded 108 weights: 182 KB.
[INFO] Prompt executed in 1.94 seconds
```

### Other

_No response_

Start with the attached Workflow.txt and reproduce repeated runs of the Latina girl image template using the reported AMD Radeon 890M, PyTorch 2.9.1 with ROCm 7.2.1, and ComfyUI 0.26.0 environment. Inspect the invalid-value warning at nodes.py:1659 and the surrounding image conversion path; done means subsequent executions produce valid images rather than the split, corrupted output.

Written by the indexing model from the issue text.