SD.Next Download: An Advanced Stable Diffusion Interface
SD.Next · for Windows, Linux and macOS
Developer: Vladimir Mandic (opens in a new tab)
- Windows
- Linux
- macOS
The download link points to the program’s official source; the file is not hosted on this site. Link address: github.com
Link not working? Let us know
SD.Next is a free, open-source web interface for generating and editing images and video with many diffusion models; it builds on the original AUTOMATIC1111 WebUI codebase. It runs on Windows, Linux and macOS with NVIDIA, AMD, Intel, Apple Silicon or CPU setups and is licensed under Apache 2.0.
SD.Next, by Vladimir Mandic, builds on the original codebase of AUTOMATIC1111’s Stable Diffusion WebUI and has grown into a separate, actively developed interface that keeps up with new image and video models. Its built-in installer manages dependencies and updates, so staying current takes little effort.
What does SD.Next do?
- Images and video: Text-to-image, image-to-image, text-to-video and image-to-video.
- Editing: Inpainting, outpainting, detailing and hires/refine passes.
- Control: LoRA, ControlNet, IP-Adapters and prompt enhancement.
- Processing: Captioning, tagging, upscaling, interpolation, colorization and filters.
- Convenience: Task queueing, extensions, multiple themes and interface languages, on desktop or mobile.
- Memory saving: CPU/GPU offloading and SDNQ quantization help larger models fit on smaller cards.
Who should use it?
It suits users who like the AUTOMATIC1111 style of interface but want support for the latest models and a wider range of hardware.
Installing SD.Next
Linux or macOS:
git clone https://github.com/vladmandic/sdnext
cd sdnext
./webui.sh
On Windows, clone the repository the same way and run webui.bat (Command Prompt) or webui.ps1 (PowerShell). The first launch can take a while as dependencies are installed. There is also an all-in-one Windows launcher (.exe) on the SD.Next Launcher releases page that includes everything it needs, plus Docker recipes for CUDA, ROCm, Intel IPEX and OpenVINO.
Supported hardware
- NVIDIA: CUDA on Windows and Linux.
- AMD: ROCm on Linux and Windows, or ZLUDA on Windows.
- Intel Arc: OneAPI and IPEX XPU on Windows and Linux.
- Apple: M1/M2 Macs via PyTorch MPS.
- Other: OpenVINO devices and CPU-only use. No fixed VRAM minimum is published; offloading and quantization reduce memory needs.
Licensing
SD.Next is free and open source under the Apache 2.0 license. Models you download have their own licenses.
Pros
- Actively developed, with support for new models
- Works with NVIDIA, AMD, Intel and Apple hardware
- Generates both images and video
Cons
- Many options make it complex for beginners
- The first launch is slow while dependencies install
Frequently Asked Questions
What is the difference between SD.Next and AUTOMATIC1111?
SD.Next builds on AUTOMATIC1111’s original codebase but is developed separately and more actively, with support for newer image and video models and more hardware back ends.
Does SD.Next work with AMD graphics cards?
Yes. It supports AMD GPUs through ROCm on Linux and Windows, and through ZLUDA on Windows.
SD.NextOpen source
Download (opens in a new tab)