Inference needs about 17 GB of GPU memory at 1024² and 29 GB at 2048², so a 24 GB card covers 1024² and 2048² needs 32 GB. Outputs are affine-invariant, i.e. depth up to an unknown scale and shift per image.
angular loss + iREPA + SinkLoss, VAE decoder fine-tuned.
albedo
linear RGB albedo in [0, 1]
L1 + iREPA, VAE decoder fine-tuned.
Log and linear depth increase with distance, disparity decreases. Each folder holds one trainables.safetensors with keys <component>.<parameter> (Diffuser.* LoRA weights, VAE.* decoder weights when present). qwen_text_embeddings/ contains the precomputed prompt embeddings and masks per modality (<prefix>_prompt_embeds.pt, <prefix>_prompt_mask.pt), so the text encoder of the base model is never needed. manifest.json lists sizes and SHA-256 checksums of all files.
Depth: Hypersim and Virtual KITTI 2 (about 74k images, as repackaged for Marigold V1). See-through depth: LayeredDepth-Syn. Normals and albedo: Hypersim. All released models were trained on a single 32 GB GPU.
@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},
year = {2026},
issue_date = {December 2026},
publisher = {Association for Computing Machinery},
volume = {45},
number = {6},
url = {https://doi.org/10.1145/3842528},
doi = {10.1145/3842528},
journal = {ACM Trans. Graph.},
month = dec,
articleno = {204},
numpages = {14}
}
The weights in this repository are released under the Apache License, Version 2.0 (see LICENSE). The base model Qwen-Image-Edit-2509 keeps its own license.
Files
33 files · 17.3 GBWeights 16.2 GB (safetensors). Sizes and sha256 from the signed manifest.
BAAI