pkuliyi2015/multidiffusion-upscaler-for-automatic1111

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

4994 stars · 348 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

An Automatic1111 Stable Diffusion WebUI extension that enables generating and upscaling ultra-large images (2K+) on GPUs with limited VRAM (≤6GB). It implements Tiled Diffusion methods (MultiDiffusion, Mixture of Diffusers, DemoFusion), an original Tiled VAE technique, and Tiled Noise Inversion.

Use cases

• generate ultra-large images with limited VRAM
• upscale images 4x with detail enhancement
• run stable diffusion on a 6GB GPU
• create panoramic images with txt2img
• control prompts per region in one image
• reduce VRAM usage during VAE decoding

When to choose

• you use Automatic1111 WebUI and need 2K+ images on consumer GPUs
• you hit out-of-memory errors on large generations or upscaling
• you need regional prompt control for multi-character scenes

When to avoid

• you use ComfyUI or another UI without Automatic1111
• you need a commercially-licensed tool (post-2023-03-28 code is CC BY-NC-SA, non-commercial)
• you only generate small images and don't need tiling overhead

Facets

plugin · maturity active

image-processing stable-diffusion llm-inference image-processing artificial-intelligence web-development python cross-platform stable-diffusion-webui automatic1111-extension tiled-diffusion tiled-vae vram-optimization upscaling multidiffusion large-image-generation gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="pkuliyi2015/multidiffusion-upscaler-for-automatic1111")