👇👀 Read me! 👇👀
Thanks for 3k followers and 10 MILLION generations on CivitAI! 🥳🎉 As a thank you, I am releasing this slider FOR FREE! Enjoy!
This is a slider for IllustriousXL designed to provide support and fine control of the "foreshortening" concept. This is a unipolar slider LoRA (only supports positive input weights).
The activation tag is foreshortening and is required for this LoRA to work well.
Safe weighting is 0 to 2 . Positive values increase/introduce the foreshortening concept.
If you generate images with this, please share! I love seeing what people make with my LoRAs and it helps keep me motivated to make more. If you have IllustriousXL LoRA requests, feel free to leave a comment and I'll look into into it 👍
Want to support my work? Check out my Ko-Fi! https://ko-fi.com/shedtheskin
Known Issues:
1.) High input values may also change the camera angle slightly more than necessary
Changelog:
September 1, 2025 - v1.0 Initial release
Description
v1.0 Initial Release
FAQ
Comments (8)
Extremely useful. Thank you.
Super fun! Keep up the good work!
I get this on Invoke AI:
Server Error
ValueError: The SDXL LoRA could only be partially converted to diffusers format. converted=2166, not_converted=1038
any idea ?
@XX0XX Does the console give you any more information? As is, the error doesn't really tell me what the problem actually is.
@Shed_The_Skin
Loading pipeline components...: 100%|███████████████████████████████████| 7/7 [00:00<00:00, 13.47it/s]
[2025-11-10 06:30:23,305]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '91b8342a-4d53-47
8d-b4d3-b1f8f7cde840:text_encoder' (CLIPTextModel) onto cuda device in 0.39s. Total model size: 234.72M
B, VRAM: 234.72MB (100.0%)
[2025-11-10 06:30:23,307]::[ModelManagerService]::INFO --> [MODEL CACHE] Loaded model '91b8342a-4d53-47
8d-b4d3-b1f8f7cde840:tokenizer' (CLIPTokenizer) onto cuda device in 0.00s. Total model size: 0.00MB, VR
AM: 0.00MB (0.0%)
[2025-11-10 06:30:23,422]::[InvokeAI]::ERROR --> Error while invoking session 8a70abc2-a24b-4d87-96e7-5
3cafea874e4, invocation dfc4ccaf-1d8d-4787-a0b2-ee9a2876d20f (sdxl_compel_prompt): The SDXL LoRA could
only be partially converted to diffusers format. converted=2166, not_converted=1038
[2025-11-10 06:30:23,422]::[InvokeAI]::ERROR --> Traceback (most recent call last):
File "\.venv\Lib\site-packages\invokeai\app\services\session_processor\session_processor_defa
ult.py", line 130, in run_node
output = invocation.invoke_internal(context=context, services=self._services)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\app\invocations\baseinvocation.py", line 244, in invo
ke_internal
output = self.invoke(context)
^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\app\invocations\compel.py", line 280, in invoke
c1, c1_pooled = self.run_clip_compel(context, self.clip, self.prompt, False, "lora_te1_", zero_on_e
mpty=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\app\invocations\compel.py", line 188, in run_clip_com
pel
LayerPatcher.apply_smart_model_patches(
File "\AppData\Roaming\uv\python\cpython-3.12.9-windows-x86_64-none\Lib\contextlib.py", lin
e 137, in enter
return next(self.gen)
^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\backend\patches\layer_patcher.py", line 39, in apply_
smart_model_patches
for patch, patch_weight in patches:
^^^^^^^
File "\.venv\Lib\site-packages\invokeai\app\invocations\compel.py", line 173, in loraloader
lora_info = context.models.load(lora.lora)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\app\services\shared\invocation_context.py", line 392,
in load
return self._services.model_manager.load.load_model(model, submodel_type)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".venv\Lib\site-packages\invokeai\app\services\model_load\model_load_default.py", line
71, in load_model
).load_model(model_config, submodel_type)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\backend\model_manager\load\load_default.py", line 57,
in load_model
cache_record = self._load_and_cache(model_config, submodel_type)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\backend\model_manager\load\load_default.py", line 78,
in loadand_cache
loaded_model = self._load_model(config, submodel_type)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\backend\model_manager\load\model_loaders\lora.py", li
ne 97, in loadmodel
state_dict = convert_sdxl_keys_to_diffusers_format(state_dict)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "\.venv\Lib\site-packages\invokeai\backend\patches\lora_conversions\sdxl_lora_conversion
utils.py", line 57, in convertsdxl_keys_to_diffusers_format
raise ValueError(
ValueError: The SDXL LoRA could only be partially converted to diffusers format. converted=2166, not_co
nverted=1038
@XX0XX Hm, doesn't look like these logs really expound on the error. if you haven't already, I'd recommend opening an issue on Invoke AI's GitHub. If you do, you can tag me @Shed-The-Skin on Github. Unfortunately, I don't have the time to really track this down myself right now.
@Shed_The_Skin its fine ill just skip it i have replacers for it I have a collection of 3919 safetensor files so its all good Just odd that some file formats wont work on New version of Invoke is all.
Killer resource, love this slider!
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.