Downloads · 30 days
0
syntverk/syntverk-image-models
syntverk-image-models is a machine learning model from syntverk. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as other.
Single-file safetensors / ckpt mirrors of the inpainting checkpoints used by Syntverk, bundled here so the app can fetch them from a single ungated location instead of relying on third-party mirrors that may disappear…
Downloads · 30 days
0
Access
Public
Updated Apr 7, 2026
Repo size
11.4 GB
Likes
0
Public
Click a slice to open those files.
.safetensors7.1 GB · 62%
From the Hugging Face model README
Single-file safetensors / ckpt mirrors of the inpainting checkpoints used by Syntverk, bundled here so the app can fetch them from a single ungated location instead of relying on third-party mirrors that may disappear or change without notice.
Each subdirectory contains the model file, the original license, a NOTICE
with attribution and provenance, and a SHA256 hash you can verify against
your download.
| Path | Origin | Format | Size | License |
|---|---|---|---|---|
sd15-inpaint/ | runwayml/stable-diffusion-inpainting (mirror via botp) | .ckpt | 4.27 GB | CreativeML Open RAIL-M |
sdxl-inpaint/ | Converted from diffusers/stable-diffusion-xl-1.0-inpainting-0.1 | .safetensors (fp16) | 6.5 GB | CreativeML Open RAIL++-M |
loras/sdxl-pixel-art/ | nerijs/pixel-art-xl | .safetensors | 163 MB | CreativeML Open RAIL-M |
loras.json is the curated LoRA catalog Syntverk fetches at
startup. It records each LoRA's metadata (triggers, weight, model_type),
expected SHA256, license info, and a fallback chain of sources (Civitai
version ID, HF URL, or our own mirror). LoRAs whose license permits
redistribution are mirrored under loras/ here; others are
referenced upstream and pinned by hash so any silent change is detected.
sd15-inpaint/sd-v1-5-inpainting.ckpt is bit-for-bit identical to the
original RunwayML release. Verified by SHA256 against the canonical hash
c6bbc15e3224e6973459ba78de4998b80b50112b0ae5b5c67113d56b4e366b19.sdxl-inpaint/sd_xl_base_1.0_inpainting_0.1.safetensors was generated
locally by running diffusers/scripts/convert_diffusers_to_original_sdxl.py
(v0.37.1) on the official diffusers/stable-diffusion-xl-1.0-inpainting-0.1
fp16 weights, with --half --use_safetensors. The conversion is fully
reproducible from the diffusers repo.sha256sum sd15-inpaint/sd-v1-5-inpainting.ckpt
sha256sum sdxl-inpaint/sd_xl_base_1.0_inpainting_0.1.safetensors
The expected hashes are recorded in each subdirectory's SHA256SUMS file.
The model weights remain under their original licenses (see each subdirectory). This repository is not authored by RunwayML, Stability AI, or the Diffusers team — it is a redistribution permitted by those licenses, with attribution. Use is subject to the use-based restrictions in each license.