You can deploy this model and start earning money today!

Wszelkiebro/
reliquary-depth-anything-v2-large

Name: reliquary-depth-anything-v2-large
Availability: InStock

onnx

reliquary

depth-estimation

depth-anything/Depth-Anything-V2-Large

quantized

cc-by-nc-4.0

Views

No views yet

reliquary-depth-anything-v2-large

Depth Anything V2 Large (ViT-L, ~335M params) ONNX export. Highest-quality tier for Reliquary's Heightmap Generator.

Source

ONNX re-export of the upstream model — original credits to the authors:

Repo: https://github.com/DepthAnything/Depth-Anything-V2
Paper: "Depth Anything V2" — Yang, Kang, Huang, Xu, Feng, Zhao, NeurIPS 2024.
Pretrained weights origin: HuggingFace depth-anything/Depth-Anything-V2-Large

License

Inherited from upstream — the Large variant is CC-BY-NC-4.0 (Creative Commons Attribution-NonCommercial). Non-commercial use only. Reliquary itself is non-commercial; downstream commercial users must either switch to the Small variant (Apache-2.0) or obtain a commercial license from the upstream authors.

Usage in Reliquary

This model is consumed by Reliquary, a browser-based PBR texture editor. The ONNX file is loaded via ONNX Runtime Web in the browser — no Python or server-side inference.

Runtime integration point: src/lib/ai/depthAnything.ts (variant picker)
Input shape / format: [N, 3, H, W] — RGB normalised, H/W multiples of 14.
Output shape / format: [N, 1, H, W] — relative depth, larger = closer.

Files

depth_anything_v2_large.onnx — ONNX graph (1.8 MB)
depth_anything_v2_large.onnx.data — external weights (1.2 GB, present because the model exceeds the ONNX protobuf size limit)
README.md — this file

Disclaimer

This is a re-export from the original PyTorch checkpoint to a browser-deployable ONNX format with dynamic input axes. The model weights, architecture, and training methodology are entirely the work of the original authors cited above. This repo is a hosting convenience for Reliquary's browser-side ML — it does NOT modify the model.

For the full conversion pipeline (including version-compat shims for PyTorch 2.6+, opset selection, dynamic-axes details), see Reliquary's PYTHON_CONVERSIONS_GUIDE.md.

Wszelkiebro/reliquary-depth-anything-v2-large