selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

Best open-weight Depth Estimation models

Estimate depth maps from a single image or video.

Depth estimation models predict a depth map from a single image or video, giving each pixel an estimated distance from the camera. Teams use them for tasks that need scene geometry from ordinary footage. Among the popular models listed here are Depth-Anything-V2-Small-hf from depth-anything, DepthPro-hf from apple, and DA3METRIC-LARGE from depth-anything. Each one can be downloaded and run on your own hardware.

--sort
22 models
depth-anything logo? GB

Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…

?BApache 2.0
34.3KJun 2024
apple logo? GB

We present a foundation model for zero-shot metric monocular depth estimation. Our model, Depth Pro, synthesizes high-resolution depth maps with unparalleled sharpness and high-frequency details.

?BApple Machine Learning Research License
11.1KOct 2024
depth-anything logo? GB

Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…

?BCC BY-NC 4.0
114.3KJun 2024
depth-anything logo≈0.6 GB

Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…

25MApache 2.0
1.5MJun 2024gguf
LiheYoung logo≈0.6 GB

Depth Anything model. It was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al. and first released in this repository.

25MApache 2.0
34.2KJan 2024

Dense Prediction Transformer (DPT) model trained on 1.4 million images for monocular depth estimation. It was introduced in the paper Vision Transformers for Dense Prediction by Ranftl et al. (2021) and first released in this repository.

?BApache 2.0
446.3KDec 2022
depth-anything logo

DA3METRIC-LARGE

depth-anything

≈1.3 GB

DA3 Metric Large model specialized for metric depth estimation in monocular settings, ideal for applications requiring real-world scale. Canonical metric depth; multiplying by focal length gives metric depth.

334MApache 2.0
446.5KNov 2025
depth-anything logo

DA3MONO-LARGE

depth-anything

≈1.3 GB

DA3 Monocular Large model for high-quality relative monocular depth estimation. Unlike disparity-based models (e.g., Depth Anything 2), it directly predicts depth, resulting in superior geometric accuracy.

334MApache 2.0
226.2KNov 2025
apple logo≈2.6 GB

This is the transformers version of DepthPro, a foundation model for zero-shot metric monocular depth estimation, designed to generate high-resolution depth maps with remarkable sharpness and fine-grained details. For the checkpoint compatible with the original codebase, please check this repo.

952MApple Machine Learning Research License
57.7KNov 2024
LiheYoung logo≈1.3 GB

Depth Anything model. It was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al. and first released in this repository.

335MApache 2.0
213.3KJan 2024
Intel logo≈1.3 GB

ZoeDepth model fine-tuned on the NYU and KITTI datasets. It was introduced in the paper ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth by Shariq et al. and first released in this repository.

345MMIT
38.2KApr 2024
LiheYoung logo≈0.7 GB

Depth Anything model. It was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al. and first released in this repository.

97MApache 2.0
117.8KJan 2024
depth-anything logo

DA3NESTED-GIANT-LARGE

depth-anything

≈4.3 GB

DA3 Nested model combining the any-view Giant model with the metric Large model for metric-scale visual geometry reconstruction. This is our recommended model that combines all capabilities.

1.7BCC BY-NC 4.0
112.8KNov 2025
tencent logo? GB

If you find DepthCrafter useful, please help ⭐ the [Github Repo] , which is important to Open-Source projects. Thanks!

?Blicense
6.6KSept 2024
depth-anything logo≈4.3 GB

DA3 Nested model combining the any-view Giant model with the metric Large model for metric-scale visual geometry reconstruction. This is our recommended model that combines all capabilities.

1.7BCC BY-NC 4.0
110.4KDec 2025
depth-anything logo

DA3-SMALL

depth-anything

≈0.6 GB

DA3 Small model for multi-view depth estimation and camera pose estimation. Efficient foundation model with unified depth-ray representation.

34MApache 2.0
197.6KNov 2025
depth-anything logo

DA3-LARGE

depth-anything

≈1.4 GB

DA3 Large model for multi-view depth estimation and camera pose estimation. Foundation model with unified depth-ray representation.

411MCC BY-NC 4.0
375.1KNov 2025
prs-eth logo≈2.4 GB

marigold-depth-v1-0 is an open-weight depth model with 865.9M parameters from prs-eth, published on Hugging Face.

866MApache 2.0
5.1KDec 2023
depth-anything logo

DA3-GIANT

depth-anything

≈3.5 GB

DA3 Giant model for multi-view depth estimation, camera pose estimation, and 3D Gaussian estimation. This is the flagship foundation model with unified depth-ray representation.

1.4BCC BY-NC 4.0
65.9KNov 2025
depth-anything logo

DA3-LARGE-1.1

depth-anything

≈1.4 GB

DA3 Large model for multi-view depth estimation and camera pose estimation. Foundation model with unified depth-ray representation.

411MApache 2.0
58.2KDec 2025
depth-anything logo≈1.3 GB

Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…

335MCC BY-NC 4.0
243.3KJun 2024

This model represents a fine-tuned version of Depth Anything V2 for indoor metric depth estimation using the synthetic Hypersim datasets.

25MUnknown licence
26.3KJul 2024