Vathography/vathography-app-models

Vathography App — model files

The depth model files the Vathography App downloads on its first run: the Marigold V2 depth model (depth/Log-stage2) in ComfyUI's file layout. They are byte-identical to the files in Comfy-Org/marigold-v2-0 (revision c304e4370a1bc677a98c906c3683ace49a6bdadd), kept here so that the app's download stays the same.

The app takes its backbone and its see-through model from Vathography/marigold-v2-see-through-comfyui. The app itself: https://vathography.com

Files

File	What it is	sha256
marigold_v2_depth_log_stage2.safetensors	the depth model (LoRA)	ec1f5649eef5ddb487b4de2222f6b62837f094c02d652843bd4fceb481037ca0
marigold_v2_depth_log_stage2_vae.safetensors	its VAE (fine-tuned decoder, Qwen-Image-Edit-2509 encoder)	0c1a95f7410678a93b59ba42b3a069f50b54ff9f6e35525699eb8fcd63ce332c
marigold_v2_depth_conditioning.safetensors	its conditioning	9f5c00a2df993136c541cdbc37bcfb81ab1ba81adaec25bfb9cbfff81edd7c9c

The same files work in ComfyUI (models/loras/, models/vae/, models/embeddings/).

Credits and license

Marigold V2 — Igor Pavlovic, Thiemo Wandel, Anton Obukhov, Luca Bartolomei, Andrey Davydov, Fabio Tosi, Matteo Poggi, Sabine Süsstrunk, Dengxin Dai (EPFL, HUAWEI Bayer Lab, University of Bologna). Paper · code · weights. Qwen-Image-Edit-2509 — the Qwen team. Model · report. ComfyUI file layout and these files — Comfy-Org.

Apache License 2.0 (LICENSE), as the original works. Free to use, including commercially. Not affiliated with or endorsed by the authors above. Details in NOTICE.

@article{pavlovic2026marigoldv2,
    author = {Pavlovic, Igor and Wandel, Thiemo and Obukhov, Anton and Bartolomei, Luca and Davydov, Andrey and Tosi, Fabio and Poggi, Matteo and S{\"u}sstrunk, Sabine and Dai, Dengxin},
    title = {Marigold V2: Revisiting Diffusion Transformers for Monocular Depth Estimation},
    journal = {ACM Trans. Graph.}, volume = {45}, number = {6}, articleno = {204}, year = {2026},
    doi = {10.1145/3842528}
}
@misc{wu2025qwenimagetechnicalreport,
    title = {Qwen-Image Technical Report},
    author = {Chenfei Wu and others},
    year = {2025}, eprint = {2508.02324}, archivePrefix = {arXiv}
}