[bug]: Z-Image Turbo models fail in Canvas when using Regional Guidance

[shanedk]

Is there an existing issue for this problem?

• I have searched the existing issues

Install method

Invoke's Launcher

Operating system

Windows

GPU vendor

Nvidia (CUDA)

GPU model

RTX 3080

GPU VRAM

12GB

Version number

6.13.0

Browser

No response

System Information

{
"version": "6.13.0",
"dependencies": {
"absl-py" : "2.4.0",
"accelerate" : "1.13.0",
"annotated-types" : "0.7.0",
"anyio" : "4.13.0",
"attrs" : "26.1.0",
"bcrypt" : "3.2.2",
"bidict" : "0.23.1",
"bitsandbytes" : "0.49.2",
"blake3" : "1.0.8",
"certifi" : "2022.12.7",
"cffi" : "2.0.0",
"charset-normalizer" : "2.1.1",
"click" : "8.4.1",
"colorama" : "0.4.6",
"coloredlogs" : "15.0.1",
"compel" : "2.1.1",
"contourpy" : "1.3.3",
"cryptography" : "48.0.0",
"CUDA" : "12.8",
"cycler" : "0.12.1",
"Deprecated" : "1.3.1",
"diffusers" : "0.37.0",
"dnspython" : "2.8.0",
"dynamicprompts" : "0.31.0",
"ecdsa" : "0.19.2",
"einops" : "0.8.2",
"email-validator" : "2.3.0",
"fastapi" : "0.118.3",
"fastapi-events" : "0.12.2",
"filelock" : "3.29.0",
"flatbuffers" : "25.12.19",
"fonttools" : "4.63.0",
"fsspec" : "2026.4.0",
"gguf" : "0.19.0",
"h11" : "0.16.0",
"httpcore" : "1.0.9",
"httptools" : "0.8.0",
"httpx" : "0.28.1",
"huggingface_hub" : "0.36.2",
"humanfriendly" : "10.0",
"idna" : "3.4",
"importlib_metadata" : "7.1.0",
"InvokeAI" : "6.13.0",
"jax" : "0.7.1",
"jaxlib" : "0.7.1",
"Jinja2" : "3.1.6",
"kiwisolver" : "1.5.0",
"MarkupSafe" : "3.0.3",
"matplotlib" : "3.10.9",
"mediapipe" : "0.10.14",
"ml_dtypes" : "0.5.4",
"mpmath" : "1.3.0",
"networkx" : "3.6.1",
"numpy" : "1.26.4",
"onnx" : "1.16.1",
"onnxruntime" : "1.19.2",
"opencv-contrib-python": "4.11.0.86",
"opt_einsum" : "3.4.0",
"packaging" : "24.1",
"passlib" : "1.7.4",
"picklescan" : "1.0.4",
"pillow" : "12.2.0",
"prompt_toolkit" : "3.0.52",
"protobuf" : "4.25.9",
"psutil" : "7.2.2",
"pyasn1" : "0.6.3",
"pycparser" : "3.0",
"pydantic" : "2.13.4",
"pydantic-settings" : "2.14.1",
"pydantic_core" : "2.46.4",
"pyparsing" : "3.3.2",
"PyPatchMatch" : "1.0.2",
"pyreadline3" : "3.5.6",
"python-dateutil" : "2.9.0.post0",
"python-dotenv" : "1.2.2",
"python-engineio" : "4.13.2",
"python-jose" : "3.5.0",
"python-multipart" : "0.0.29",
"python-socketio" : "5.16.2",
"PyWavelets" : "1.9.0",
"PyYAML" : "6.0.3",
"regex" : "2026.5.9",
"requests" : "2.28.1",
"rsa" : "4.9.1",
"safetensors" : "0.7.0",
"scipy" : "1.17.1",
"semver" : "3.0.4",
"sentencepiece" : "0.2.0",
"setuptools" : "70.2.0",
"simple-websocket" : "1.1.0",
"six" : "1.17.0",
"sounddevice" : "0.5.5",
"spandrel" : "0.4.2",
"starlette" : "0.48.0",
"sympy" : "1.14.0",
"tokenizers" : "0.22.2",
"torch" : "2.7.1+cu128",
"torchsde" : "0.2.6",
"torchvision" : "0.22.1+cu128",
"tqdm" : "4.66.5",
"trampoline" : "0.1.2",
"transformers" : "4.57.6",
"typing-inspection" : "0.4.2",
"typing_extensions" : "4.15.0",
"urllib3" : "1.26.13",
"uvicorn" : "0.48.0",
"watchfiles" : "1.2.0",
"wcwidth" : "0.7.0",
"websockets" : "16.0",
"wrapt" : "2.2.1",
"wsproto" : "1.3.2",
"zipp" : "3.19.2"
},
"config": {
"schema_version": "4.0.3",
"legacy_models_yaml_path": null,
"host": "127.0.0.1",
"port": 9090,
"allow_origins": [],
"allow_credentials": true,
"allow_methods": [""],
"allow_headers": [""],
"ssl_certfile": null,
"ssl_keyfile": null,
"log_tokenization": false,
"patchmatch": true,
"models_dir": "models",
"convert_cache_dir": "models\.convert_cache",
"download_cache_dir": "models\.download_cache",
"legacy_conf_dir": "configs",
"db_dir": "databases",
"outputs_dir": "W:\AI\InvokeAI\outputs",
"image_subfolder_strategy": "date",
"custom_nodes_dir": "nodes",
"style_presets_dir": "style_presets",
"workflow_thumbnails_dir": "workflow_thumbnails",
"log_handlers": ["console"],
"log_format": "color",
"log_level": "info",
"log_sql": false,
"log_level_network": "warning",
"use_memory_db": false,
"dev_reload": false,
"profile_graphs": false,
"profile_prefix": null,
"profiles_dir": "profiles",
"max_cache_ram_gb": null,
"max_cache_vram_gb": null,
"log_memory_usage": false,
"model_cache_keep_alive_min": 0,
"device_working_mem_gb": 3,
"enable_partial_loading": true,
"keep_ram_copy_of_weights": true,
"ram": null,
"vram": null,
"lazy_offload": true,
"pytorch_cuda_alloc_conf": "backend:cudaMallocAsync",
"device": "cuda",
"precision": "auto",
"sequential_guidance": false,
"attention_type": "torch-sdp",
"attention_slice_size": "auto",
"force_tiled_decode": false,
"pil_compress_level": 1,
"max_queue_size": 10000,
"clear_queue_on_startup": true,
"max_queue_history": null,
"allow_nodes": null,
"deny_nodes": null,
"node_cache_size": 512,
"hashing_algorithm": "random",
"remote_api_tokens": [
{"url_regex": "civitai.com", "token": "REDACTED"},
{"url_regex": "huggingface.co", "token": "REDACTED"},
{"url_regex": "civitai.red", "token": "REDACTED"}
],
"scan_models_on_startup": false,
"unsafe_disable_picklescan": false,
"allow_unknown_models": true,
"multiuser": false,
"strict_password_checking": false,
"external_alibabacloud_api_key": null,
"external_alibabacloud_base_url": null,
"external_gemini_api_key": null,
"external_openai_api_key": null,
"external_gemini_base_url": null,
"external_openai_base_url": null,
"external_seedream_api_key": null,
"external_seedream_base_url": null
},
"set_config_fields": [
"hashing_algorithm", "attention_type", "clear_queue_on_startup", "legacy_models_yaml_path",
"device", "remote_api_tokens", "pytorch_cuda_alloc_conf", "outputs_dir",
"image_subfolder_strategy"
]
}

What happened

When using Regional Guidance layers on a Canvas, selecting a Z-Image Turbo model causes the following error:

[2026-05-29 17:55:22,968]::[InvokeAI]::ERROR --> Error while invoking session c29f1848-cb22-4c05-9793-37764718cd15, invocation 69e07761-52c9-489a-8151-82249feeb595 (z_image_denoise): split_with_sizes e
xpects split_sizes to sum exactly to 46 (input tensor's size at dimension 0), but got split_sizes=[32]
[2026-05-29 17:55:22,968]::[InvokeAI]::ERROR --> Traceback (most recent call last):
File "W:\AI\InvokeAI.venv\Lib\site-packages\invokeai\app\services\session_processor\session_processor_default.py", line 130, in run_node
output = invocation.invoke_internal(context=context, services=self._services)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\invokeai\app\invocations\baseinvocation.py", line 244, in invoke_internal
output = self.invoke(context)
^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\torch\utils_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\invokeai\app\invocations\z_image_denoise.py", line 130, in invoke
latents = self._run_diffusion(context)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\invokeai\app\invocations\z_image_denoise.py", line 721, in _run_diffusion
model_output = transformer(
^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\invokeai\backend\z_image\z_image_transformer_patch.py", line 228, in
transformer.forward = lambda *args, **kwargs: regional_fwd(transformer, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\invokeai\backend\z_image\z_image_transformer_patch.py", line 103, in regional_forward
cap_freqs_cis = list(self.rope_embedder(torch.cat(cap_pos_ids, dim=0)).split(cap_item_seqlens, dim=0))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "W:\AI\InvokeAI.venv\Lib\site-packages\torch_tensor.py", line 1053, in split
return torch._VF.split_with_sizes(self, split_size, dim)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: split_with_sizes expects split_sizes to sum exactly to 46 (input tensor's size at dimension 0), but got split_sizes=[32]

What you expected to happen

By all appearances, this is a bug, especially as it works with FLUX.2 Klein and Anima models. There's no indication in the interface that this isn't supported with ZiT. If it isn't, and this isn't an actual bug to be fixed, then either a more useful toast should be presented to the user, or the Regional Guidance layers could simply be disabled when a ZiT model is selected.

How to reproduce the problem

1. Do Canvas stuff
2. Make at least one Regional Guidance layer
3. Select a Z-Image Turbo model
4. Hit Invoke

Additional context

No response

Discord username

shanedk