ComfyUI Extension: ComfyUI-DepthAnythingV3
ComfyUI nodes to use Depth Anything V3 - the latest depth estimation model from ByteDance. For now monocular depth, camera pose estimation and 3D point clouds/gaussians are supported. Models autodownload from HuggingFace (depth-anything org). This is a first draft, let me know if you have any feedback! :)
Custom Nodes (0)
README
ComfyUI Depth Anything V3
Custom nodes for Depth Anything V3 integration with ComfyUI.
Description
Depth Anything V3 is the latest depth estimation model that predicts spatially consistent geometry from visual inputs.
Published: November 14, 2025 Paper: Depth Anything 3: Recovering the Visual Space from Any Views
Nodes
Download And Load Depth Anything V3 Model - Downloads and loads model from HuggingFace
- Inputs:
model(variant),precision(auto/fp16/fp32/bf16) - Output:
da3_model - Models auto-download to
ComfyUI/models/depthanything3/
Depth Anything V3 - Runs depth estimation
- Inputs:
da3_model,images - Output:
depth(normalized 0-1, grayscale)
Model Variants
| Model | Size | Features | |-------|------|----------| | DA3-Small | 80M | Fast, good quality | | DA3-Large | 350M | High quality, balanced | | DA3-Giant | 1.15B | Best quality, slower | | DA3Mono-Large | 350M | Optimized for monocular depth | | DA3Metric-Large | 350M | Metric depth estimation | | DA3Nested-Giant-Large | 1.4B | Combined model with metric scaling |
Credits
- Original Paper: Haotong Lin, Sili Chen, Jun Hao Liew, et al. (ByteDance Seed Team)
- Implementation: Based on the official Depth Anything 3 repository
- Inspiration: kijai's ComfyUI-Depth-Anything-V2
License
Model architecture files based on Depth Anything 3 (Apache 2.0 / CC BY-NC 4.0 depending on model).
Note: Some models (Giant, Nested) use CC BY-NC 4.0 license (non-commercial use only).