Image & Vision AGPL-3

Automatic1111 (A1111)

The original Stable Diffusion WebUI. Most users should migrate to Forge, but A1111 still has the largest extension ecosystem for legacy workflows.

Website
Platforms: windowsmacoslinux

Automatic1111 (universally “A1111”) is the Stable Diffusion WebUI that defined the local image-generation scene in 2022-2023. It wrapped CompVis’s inference script behind a Gradio interface, added txt2img/img2img/inpaint/outpaint, and shipped an extension system that let the community bolt on everything from ControlNet to LoRA managers to latent couple. For two years it was the answer to “how do I run Stable Diffusion locally?” In 2026 its role has shifted: Forge absorbed A1111’s UI but rewrote the backend for speed and low-VRAM cards, and most active development has moved there. A1111 itself is now in maintenance mode. It still matters because of its extension library — hundreds of plugins that never made the jump to Forge — and because so many tutorials, Civitai workflows, and muscle-memory habits are written against it.

Key Features

  • The canonical Gradio UI. Tabs for txt2img, img2img, inpaint, outpaint, extras, checkpoints, LoRAs, hypernets, embeddings — the layout every other SD UI is compared against. If you learned Stable Diffusion anywhere, you learned it here.
  • Largest legacy extension library. The Extensions tab can install from a URL or browse the built-in index. ControlNet, ADetailer, Regional Prompter, OpenPose editor, ReActor face swap, and dozens of LoRA/embedding managers originated as A1111 extensions.
  • Prompt syntax that became a standard. Weighting (word:1.3), alternation [a|b], andBREAK-style fragmentation are A1111 conventions that other tools emulate for compatibility.
  • Mature, documented, predictable. No surprises in the UI; every setting has a tooltip and a three-year trail of forum threads explaining it.

When to Use Automatic1111 (A1111)

Use A1111 when an extension you depend on has not been ported to Forge, when you are following an older tutorial step-by-step, or when you maintain a working pipeline and have no reason to migrate. For everything else — especially SDXL on 6-8GB cards, Flux, or any new architecture — switch to Forge or ComfyUI. Treat A1111 as the legacy compatibility option, not the default.

Setup

git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
cd stable-diffusion-webui
# Windows: run webui-user.bat (sets PYTHON, VENV, COMMANDLINE_ARGS)
# Linux/macOS:
./webui.sh

# Common flags in webui-user.sh / webui-user.bat:
#   export COMMANDLINE_ARGS="--xformers --medvram --api --listen"
#   export PYTHON="/usr/bin/python3.10"

Python 3.10.x is the supported version; 3.11+ can break older extensions. Checkpoints go in models/Stable-diffusion, LoRAs in models/Lora. The first run builds a venv and downloads dependencies, which can take 10-20 minutes.

See also