`
- see [comment](https://github.com/Haoming02/sd-webui-forge-classic/issues/1439#issuecomment-5539099537) for example
#### Removed Features
- [X] SD2
- [X] SD3
- [X] Forge Spaces
- [X] Hypernetworks
- [X] CLIP Interrogator
- [X] Deepbooru Interrogator
- [X] Textual Inversion Training
- [X] Some built-in Extensions
- [X] Some built-in Scripts
- [X] Some Samplers & Schedulers
- [X] Some Compatibility Settings
- [X] Stealth Infotext
- [X] `bitsandbytes` Support
- [ ] `Nunchaku` Support
#### Optimizations
- [X] **[Comfy]** Rewrite the Backend *(`memory_management.py`, `ModelPatcher`, `attention.py`, etc.)*
- [X] No longer `git` `clone` any repository on fresh install
- [X] No longer create the UI twice when using `--api`
- [X] No longer install `open-clip`
- [X] Fix memory leak when switching checkpoints
- [X] Restore the ability to drag-and-drop images onto `gr.Image` that already contains image
- [X] Speed up launch time
- [X] Speed up model loading
- [X] Improve timer logs
- [X] Remove unused `cmd_args`
- [X] Remove unused `args_parser`
- [X] Remove unused `shared_options`
- [X] Remove legacy codes
- [X] Fix some typos
- [X] Fix automatic `Tiled VAE` fallback
- [X] Fix `Tiling` for SD1 and SDXL
- [X] Pad conditioning for SDXL
- [X] Remove duplicated upscaler codes
- [X] Update [spandrel](https://github.com/chaiNNer-org/spandrel)
- support new upscaler architectures
> [!Note]
> Put every upscaler (`.pth` / `.safetensors`) inside the `ESRGAN` folder
> [!Tip]
> Check out [OpenModelDB](https://openmodeldb.info/) for where to get upscalers
- [X] Improve `ForgeCanvas`
- brush adjustments
- customization
- deobfuscate
- eraser
- hotkeys
- mobile friendly
- [X] Optimize upscaler logics
- [X] Optimize certain operations in `Spandrel`
- [X] Optimize certain operations for `VAE`
- [X] Speed up model loading
- [X] Streamline model sorting
- [X] Improve color correction
- [X] Improve memory management
- [X] Unload all models on `OutOfMemory`
- [X] Update the implementation for `X/Y/Z Plot`
- [X] Update the implementation for `Soft Inpainting`
- [X] Update the implementation for `MultiDiffusion`
- [X] Update the implementation for `uni_pc` and `LCM` samplers
- [X] Update the implementation of LoRAs
- [X] Streamline Model Refresh
- clicking refresh in Quicksettings will now also update `Hires. fix`, `Refiner`, and `X/Y/Z Plot` dropdowns
- [X] Revamp settings
- improve formatting
- update descriptions
- [X] Check for Extension updates in parallel
- [X] Move `embeddings` folder into `models` folder
- [X] Infotext Rewrite
- allow switching Models and Modules
- save `emphasis` properly
- correct default values
- [X] ControlNet Rewrite
- change Units to `gr.Tab`
- improve `masks` & `buttons`
- remove multi-inputs, as they are "[misleading](https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/932)"
- [X] Checkpoint Merger Rewrite
- update ui
- support modern models
- [X] Refiner Rewrite
- update ui
- improve tooltips
- implement `Refiner CFG Scale`
- [X] Improve inference speed
- [X] Improve non-Nvidia support
- [X] Lint & Format
- [X] Update `Pillow`
- faster image processing
- [X] Update `protobuf`
- faster `insightface` loading
- [X] Update to latest PyTorch
- `torch==2.14.0+cu130`
> [!Note]
> If your GPU does not support the latest PyTorch, manually [install](https://github.com/Haoming02/sd-webui-forge-classic/wiki/Extra-Installations#older-pytorch) older version of PyTorch
- [X] Update some packages to newer versions
- [X] Update recommended Python to `3.13.15`
- [X] many more... :tm:
## Commandline
> These flags can be added after the `set COMMANDLINE_ARGS=` line in the `webui-user.bat` *(in the same line ; separate each flag with space)*
> [!Tip]
> Use `python launch.py --help` to see all available flags
- `--xformers`: Install the `xformers` package to speed up generation
> [!Warning]
> `xformers` does **not** support `RTX 50s`
- `--port`: Specify a server port to use
- defaults to `7860`
- `--api`: Enable [API](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/API) access
#### by. Neo
- `--cuda-malloc`: Improve memory allocation
- `--cuda-stream`: Enable async weight offloading
- `--pin-shared-memory`: Improve RAM utilization
- `--expandable-segments`: Enable experimental PyTorch allocator *(may prevent `OutOfMemory` errors on certain platforms)*
- `--uv`: Replace the `python -m pip` calls with `uv pip` to massively speed up package installation
- require **uv** to be installed first *(see [Extra Installations](https://github.com/Haoming02/sd-webui-forge-classic/wiki/Extra-Installations))*
- `--uv-symlink`: Same as above; but additionally pass `--link-mode symlink` to the commands
- significantly reduce installation size (`~7 GB` to `~100 MB`)
- `--uv-local-cache`: Same as above; but additionally set `UV_CACHE_DIR` to a `.uv-cache` folder within WebUI directory
- speed up installation on non-default drive *(**i.e.** not `C:` on Windows)*
- allow clean uninstallation by simply deleting the WebUI directory
> [!Important]
> `symlink` means it will directly access the packages from the cache folder instead of copying the packages over ; refrain from clearing the cache when using this option
- `--model-ref`: Points to a central `models` folder that contains all your models
- said folder should contain subfolders like `Stable-diffusion`, `Lora`, `VAE`, `ESRGAN`, etc.
> [!Important]
> This simply **replaces** the `models` folder rather than adding on top of it
- `--forge-ref-a1111-home`: Point to an Automatic1111 installation to load its `models` folders
- **i.e.** `Stable-diffusion`, `text_encoder`, etc.
- `--forge-ref-comfy-home`: Point to a ComfyUI installation to load its `models` folders
- **i.e.** `diffusion_models`, `clip`, etc.
- `--forge-ref-comfy-yaml`: Point to the ComfyUI `extra_model_paths.yaml` to load its configurations
- **i.e.** `base_path`, `checkpoints`, etc.
- `--sage`: Install the `sageattention` package to speed up generation
- will also attempt to install `triton` automatically
- `--flash`: Install the `flash_attn` package to speed up generation
- `--nunchaku`: Install the `nunchaku` package to inference SVDQ models
- `--fast-fp8`: Use the `torch._scaled_mm` function when the model type is `float8_e4m3fn`
- `--fast-fp16`: Enable the `allow_fp16_accumulation` option
- `--autotune`: Enable the `torch.backends.cudnn.benchmark` option
- this is slower in my experience...
- `--tiled-conv2d`: Replace `Conv2d` ops with tiled variants
- has greater reduction for **SD1** and **SDXL** VAE; less for **Wan** VAE
- `64` / `128` / `256` / `512`
- `--disable-preset`: Remove the specified Preset(s) from the dropdown and settings
- `--sort-preset`: Sort the Preset dropdown alphabetically
- `--use-ck-attention`: Use the Comfy-Kitchen attention to speed up generation
- override other attention packages
- `--enable-triton-backend`: Enable the use of Triton backend in `comfy-kitchen`
- *might* speed up inference
- `--force-w4a8`: Always load `convrot_w4a4` models in `w4a8` format instead
- improve quality
- `--pynvml`: Use the NVIDIA Management Library to determine the actual amount of free VRAM
- prevent system freeze when running other programs
## Installation
0. Install **[git](https://git-scm.com/downloads)**
1. Clone the Repo
```bash
git clone https://github.com/Haoming02/sd-webui-forge-classic sd-webui-forge-neo --branch neo
```
2. Setup Python
Recommended Method
- Install **[uv](https://github.com/astral-sh/uv#installation)**
- Set up **venv**
```bash
cd sd-webui-forge-neo
uv venv venv --python 3.13 --seed
```
- Add the `--uv` flag to `webui-user.bat`
Deprecated Method
- Install **[Python 3.13.12](https://www.python.org/downloads/release/python-31312/)**
- Remember to enable `Add Python to PATH`
3. **(Optional)** Configure [Commandline](#commandline)
4. Launch the WebUI via `webui-user.bat`
5. During the first launch, it will automatically install all the requirements
6. Once the installation is finished, the WebUI will start in a browser automatically
> [!Tip]
> - For **AMD**, refer to CS1o 's Guide
> - For **Linux** and **macOS**, refer to [Wiki](https://github.com/Haoming02/sd-webui-forge-classic/wiki/Unix)
> - For **Docker** (`Nvidia` / `AMD`), refer to [Docker](docker/)
> [!Tip]
> Check out [Extra Installations](https://github.com/Haoming02/sd-webui-forge-classic/wiki/Extra-Installations) for how to install `git`, `uv`, and `FFmpeg`
## Attention Functions
> [!Note]
> The `--xformers`, `--flash`, and `--sage` args are only responsible for installing the packages, **not** whether its respective attention is used *(this also means you can remove them once the packages are successfully installed)*
> [!Caution]
> Do **not** just blindly install all of them
> Nowadays the native PyTorch `scaled_dot_product_attention` is usually as fast, and also more stable
**Forge Neo** tries to import the packages and automatically choose the first available attention function in the following order:
1. `SageAttention`
2. `FlashAttention`
3. `xformers`
4. `PyTorch`
5. `Basic`
> [!Tip]
> To skip a specific attention, add the respective disable arg such as `--disable-sage`
## Issues & Requests
- **Issues** about removed features will simply be ignored
- **Issues** that is obviously user-error will simply be ignored
- **Issues** regarding **AMD** GPU will simply be ignored
- **Issues** running non-official models will simply be ignored
- do not just randomly download every single finetune/quant you find
- **Issues** about 3rd-party Extensions will simply be ignored
- extension should support the UI, not the other way around
- **Issues** caused by [StabilityMatrix](https://github.com/LykosAI/StabilityMatrix) will simply be ignored
- only open an Issue if you can reproduce it on a clean install following the official [Installation](#installation) instruction
> [!Caution]
> - If you post **NSFW** images/videos, you will immediately be banned
> - the sole discretion is on me ; if you are unsure, just generate `cats` and `dogs`...
> [!Tip]
> Check out the [Wiki](https://github.com/Haoming02/sd-webui-forge-classic/wiki) & [FAQ](https://github.com/Haoming02/sd-webui-forge-classic/issues/414)
Special thanks to AUTOMATIC1111, lllyasviel, and comfyanonymous, kijai, city96,
along with the rest of the contributors,
for their invaluable efforts in the open-source image generation community
Buy me a Coffee ☕~
PayPal me 💳~