Repository navigation
[Bug]: deforum has broken forge neo #24
Description
What happened and why was it ungood?
My forge neo was working well, then I tried to install sd-forge-deforum.
I followed the instruction (git clone, setup.sh, installed dependencies). I skipped downloading models because command didnt work. I tried launching forge, and this is what I get now :
Traceback (most recent call last):
File "sd-webui-forge-neo\venv\Lib\site-packages\diffusers\utils\import_utils.py", line 1016, in _get_module
return importlib.import_module("." + module_name, self.__name__)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "Users\g\AppData\Roaming\uv\python\cpython-3.11.14-windows-x86_64-none\Lib\importlib\__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1147, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 690, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 940, in exec_module
File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
File "sd-webui-forge-neo\venv\Lib\site-packages\diffusers\loaders\ip_adapter.py", line 40, in <module>
from ..models.attention_processor import (
File "sd-webui-forge-neo\venv\Lib\site-packages\diffusers\models\attention_processor.py", line 35, in <module>
import xformers.ops
File "sd-webui-forge-neo\venv\Lib\site-packages\xformers\ops\__init__.py", line 9, in <module>
from .fmha import (
File "sd-webui-forge-neo\venv\Lib\site-packages\xformers\ops\fmha\__init__.py", line 10, in <module>
from . import (
File "sd-webui-forge-neo\venv\Lib\site-packages\xformers\ops\fmha\flash3.py", line 110, in <module>
from ...flash_attn_3 import _C # type: ignore[attr-defined] # noqa: F401
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ImportError: DLL load failed while importing _C: The specified module could not be found.
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "sd-webui-forge-neo\launch.py", line 54, in <module>
main()
File "sd-webui-forge-neo\launch.py", line 50, in main
start()
File "sd-webui-forge-neo\modules\launch_utils.py", line 512, in start
import webui
File "sd-webui-forge-neo\webui.py", line 23, in <module>
initialize.imports()
File "sd-webui-forge-neo\modules\initialize.py", line 37, in imports
from modules import gradio_extensions, processing, ui # noqa: F401
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "sd-webui-forge-neo\modules\processing.py", line 22, in <module>
import modules.sd_models as sd_models
File "sd-webui-forge-neo\modules\sd_models.py", line 11, in <module>
from backend.loader import forge_loader
File "sd-webui-forge-neo\backend\loader.py", line 12, in <module>
from backend.diffusion_engine.chroma import Chroma
File "sd-webui-forge-neo\backend\diffusion_engine\chroma.py", line 6, in <module>
from backend.modules.k_prediction import PredictionFlux
File "sd-webui-forge-neo\backend\modules\k_prediction.py", line 6, in <module>
from diffusers.pipelines.flux.pipeline_flux import calculate_shift
File "sd-webui-forge-neo\venv\Lib\site-packages\diffusers\pipelines\flux\pipeline_flux.py", line 30, in <module>
from ...loaders import FluxIPAdapterMixin, FluxLoraLoaderMixin, FromSingleFileMixin, TextualInversionLoaderMixin
File "<frozen importlib._bootstrap>", line 1229, in _handle_fromlist
File "\sd-webui-forge-neo\venv\Lib\site-packages\diffusers\utils\import_utils.py", line 1006, in __getattr__
module = self._get_module(self._class_to_module[name])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "sd-webui-forge-neo\venv\Lib\site-packages\diffusers\utils\import_utils.py", line 1018, in _get_module
raise RuntimeError(
RuntimeError: Failed to import diffusers.loaders.ip_adapter because of the following error (look up to see its traceback):
DLL load failed while importing _C: The specified module could not be found.
and forge won't start anymore
Activity
fix: Make Depth-Anything V3 optional to prevent xformers/flash-attn c…
docs: Add comprehensive fix guide for issue #24
Tok commentedon Jan 29, 2026
✅ Fixed in commits 3070acd and 10700bc
Thank you @dees-sp for reporting this critical issue! You've helped identify a dependency conflict that was breaking Forge Neo installations for users.
Root Cause
The extension was auto-installing Depth-Anything V3, which has optional dependencies (xformers, flash-attn) that require compiled CUDA extensions. When these fail to install properly on systems without CUDA toolkit or on Python 3.12+, they create partial installations that break Forge's entire import chain.
What's Fixed
1. requirements.txt changes:
diffusers>=0.36.0,<0.37.0(matches Forge Neo 0.36.0)transformers>=4.36.0,<=4.56.2(matches Forge Neo 4.56.2)2. New documentation:
docs/TROUBLESHOOTING.md- Comprehensive troubleshooting guidedocs/ISSUE_24_FIX.md- Step-by-step repair instructionsCLAUDE.mdandsetup.shwith dependency notesHow to Fix Your Installation
Quick Repair (5 minutes)
Forge should now launch successfully! 🎉
Detailed Guide
See
docs/ISSUE_24_FIX.mdfor comprehensive repair instructions, troubleshooting, and clean reinstall options.What Still Works (Everything!)
✅ All animation modes:
✅ All depth features:
✅ All video features:
✅ All models:
What Requires Manual DA3 Install (Advanced Features Only)
❌ Only if you need:
To install DA3 (optional):
# Only if you have Python 3.11 + CUDA toolkit! pip install git+https://github.com/ByteDance-Seed/Depth-Anything-3.gitImpact
Testing
deforum/depth/depth.py:117-128Thanks again for the detailed bug report! This fix will prevent many users from experiencing broken installations. If you encounter any issues with the repair steps, please let us know.
Additional resources:
docs/TROUBLESHOOTING.mddocs/ISSUE_24_FIX.mdfeat(flux2): Add Flux 2 Klein model support and download script
dees-sp commentedon Jan 30, 2026
wow, you're super fast! Glad I could contribute a little, thanks for your your work