Stable Diffusion Inpainting Mirror (A.I.M.I)

Two community single-file safetensors conversions of Stable Diffusion inpainting models, re-hosted for stable URLs inside the A.I.M.I desktop product. Contents are unmodified.

Inpainting models mask-fill regions of an image with prompt-guided generation — used by A.I.M.I's Edit-tab "Inpainting" workflow. The SDXL version is the default at 16 GB VRAM; SD 1.5 is the lightweight fallback.

Files

File	Upstream	Size	Purpose
sd_xl_base_1.0_inpainting_0.1.safetensors	wangqyqq/... (single-file conv of diffusers/stable-diffusion-xl-1.0-inpainting-0.1)	~6.5 GB	SDXL Inpainting 1024² native (default at 16 GB+)
sd-v1-5-inpainting.safetensors	webui/stable-diffusion-inpainting	~4.0 GB	SD 1.5 Inpainting 512² native (lightweight fallback)

License

Both models are licensed under variants of the CreativeML Open RAIL-M family:

• SDXL Inpainting 1.0: CreativeML Open RAIL++-M (Stability AI)
• SD 1.5 Inpainting: CreativeML Open RAIL-M (RunwayML / Stability AI)

Both licenses permit commercial use but attach "Use-Based Restrictions" (Attachment A) — harmful content generation prohibitions that flow through to all downstream users. Redistribution requires passing these restrictions along. See CreativeML Open RAIL-M full text and the RAIL++ extensions for SDXL.

Redistributed unchanged under the same terms.

Attribution

• SDXL: Stability AI — SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis (2023).
• SDXL Inpainting 0.1: Fine-tune by Diffusers/Stability AI team.
• SD 1.5: Rombach, Blattmann, Lorenz, Esser, Ommer — High-Resolution Image Synthesis with Latent Diffusion Models (CVPR 2022).
• SD 1.5 Inpainting: RunwayML fine-tune of SD 1.5 base.