Stable Diffussion WebUI Forge Neo - Flux Error

[MusoCity]

Package

Stable Diffusion WebGU Forge Neo

When did the issue occur?

Running the Package

What GPU / hardware type are you using?

NVidia GForce RTX 3060 12GB VRAM 32GB Ram i5 3470 CPU 3.2GHz

What happened?

getting: ValueError: Failed to find available model...
using: flux1-dev-bnb-nf4-v2.safetensors ae.safetensors
they both show in dropdown menu

Console output

Python 3.13.12 (main, Feb 3 2026, 22:53:26) [MSC v.1944 64 bit (AMD64)]
Version: neo
Launching Web UI with arguments: --skip-python-version-check --gradio-allowed-path 'E:\Stability Matrix\Data\Images'
Total VRAM 12287 MB, total RAM 32710 MB memory_management.py :: INFO
PyTorch Version: 2.11.0+cu130 memory_management.py :: INFO
VRAM State: NORMAL_VRAM memory_management.py :: INFO
Device: NVIDIA GeForce RTX 3060 (cuda:0) - memory_management.py :: INFO
native
Hint: your device supports --cuda-malloc for memory_management.py :: WARNING
potential speed improvements
Using PyTorch Cross Attention attention.py :: INFO
Using PyTorch Attention for VAE attention.py :: INFO
ControlNet preprocessor location: E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\models\ControlNetPreprocessor
ControlNet UI callback registered. controlnet_ui_group.py :: INFO
Checkpoint flux1-dev-bnb-nf4-v2.safetensors not found; loading fallback sd\flux1-dev-bnb-nf4-v2.safetensors
Model Selected: main_entry.py :: INFO
{
"checkpoint": "flux1-dev-bnb-nf4-v2.safetensors",
"modules": [
"ae.safetensors"
],
"dtype": "[torch.float16, torch.bfloat16]"
}
Patch LoRAs on-the-fly: False main_entry.py :: INFO
Running on local URL: http://127.0.0.1:7860

To create a public link, set share=True in launch().
Startup time: 32.3s (prepare environment: 5.3s, launcher: 0.8s, forge init: 8.5s, shared init: 0.3s, misc. imports: 8.8s, load scripts: 2.8s, create ui: 4.1s, gradio launch: 1.7s).
Loading Model: {'checkpoint_info': {'filename': 'E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\models\Stable-diffusion\sd\flux1-dev-bnb-nf4-v2.safetensors', 'hash': 'f0770152'}, 'additional_modules': ['E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\models\VAE\ae.safetensors'], 'unet_storage_dtype': None}
Using Detected T5XXL Data Type: torch.float8_e4m3fn loader.py :: INFO
Using Detected Model Data Type: nf4 (pre-quant) loader.py :: INFO
forge_loader: AssertionError
Traceback (most recent call last):
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\modules\sd_models.py", line 344, in forge_model_reload
sd_model = forge_loader(state_dict, additional_state_dicts=additional_state_dicts)
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\venv\Lib\site-packages\torch\utils_contextlib.py", line 124, in decorate_context
return func(*args, **kwargs)
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\backend\loader.py", line 832, in forge_loader
component = load_huggingface_component(estimated_config, component_name, lib_name, cls_name, local_path, component_sd)
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\backend\loader.py", line 424, in load_huggingface_component
with using_forge_operations(**to_args, manual_cast_enabled=need_manual_cast, bnb_dtype=storage_dtype):
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\Stability Matrix\Data\Assets\Python\cpython-3.13.12-windows-x86_64-none\Lib\contextlib.py", line 141, in enter
return next(self.gen)
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\backend\operations.py", line 880, in using_forge_operations
assert memory_management.bnb_enabled(), 'Install the "bitsandbytes" package with --bnb'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
AssertionError: Install the "bitsandbytes" package with --bnb

Failed to load diffusion model... (check README for sd_models.py :: ERROR
supported models)
Traceback (most recent call last):
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\modules_forge\main_thread.py", line 31, in work
self.result = self.func(*self.args, **self.kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\modules\txt2img.py", line 136, in txt2img_function
processed = processing.process_images(p)
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\modules\processing.py", line 824, in process_images
manage_model_and_prompt_cache(p)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\modules\processing.py", line 793, in manage_model_and_prompt_cache
p.sd_model, just_reloaded = forge_model_reload()
~~~~~~~~~~~~~~~~~~^^
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\venv\Lib\site-packages\torch\utils_contextlib.py", line 124, in decorate_context
return func(*args, **kwargs)
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\modules\sd_models.py", line 351, in forge_model_reload
raise BufferError("Failed to load diffusion model...") from None
BufferError: Failed to load diffusion model...

Version

2.15.7.0

What Operating System are you using?

Windows

[NeuralFault]

as per the error log:
File "E:\Stability Matrix\Data\Packages\Stable Diffusion WebUI Forge - Neo\backend\operations.py", line 880, in using_forge_operations
assert memory_management.bnb_enabled(), Install the "bitsandbytes" package with --bnb
AssertionError: Install the "bitsandbytes" package with --bnb

Failed to load diffusion model... (check README for sd_models.py :: ERROR
supported models)

And per Forge Neo's Readme: "--bnb: Install the bitsandbytes package to do low-bits (nf4) inference"

Go to the launch settings (gear icon) in SM packages tab to open the launch options for Forge Neo.
Then at the bottom of the launch options window there is "Extra Launch Arguments".
Add --bnb to the input field there and click Save.

Then launch Forge Neo once and let it go through the process of installing bitsandbytes. The model should work then.
After that launch, can remove the launch argument you added for future launches (unless the bitsandbytes will need to be upgraded at some point, which you will know when the error comes up).