--- title: Depth-Anything-V3 canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/Depth-Anything-V3" md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/Depth-Anything-V3.md" repository: Qualcomm-AI-Hub/Depth-Anything-V3 last_updated: 2026-10-16 license: apache-2.0 downloads: 0 stars: 0 tags: - android --- # Depth-Anything-V3 > Depth-Anything-V3 - Qualcomm-AI-Hub 在 ModelScope 开源的模型。Depth-Anything-V3: Optimized for Qualcomm Devices Qualcomm-AI-Hub/Depth-Anything-V3 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。 - **Repository**: Qualcomm-AI-Hub/Depth-Anything-V3 - **License**: apache-2.0 - **Tags**: android - **Downloads**: 0 - **Stars**: 0 - **Last updated**: 2026-10-16 Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/Depth-Anything-V3 --- ![](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/depth_anything_v3/web-assets/model_demo.png) # Depth-Anything-V3: Optimized for Qualcomm Devices Depth Anything 3 (DA3), a model that predicts spatially consistent geometry from arbitrary visual inputs, with or without known camera poses. This is based on the implementation of Depth-Anything-V3 found [here](https://github.com/ByteDance-Seed/Depth-Anything-3). This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/depth_anything_v3) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary). Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.cn) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device. ## Getting Started There are two ways to deploy this model on your device: ### Option 1: Download Pre-Exported Models Below are pre-exported model assets ready for deployment. | Runtime | Precision | Chipset | SDK Versions | Download | |---|---|---|---|---| | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/depth_anything_v3/releases/v0.62.1/depth_anything_v3-onnx-float.zip) | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/depth_anything_v3/releases/v0.62.1/depth_anything_v3-qnn_dlc-float.zip) | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/depth_anything_v3/releases/v0.62.1/depth_anything_v3-tflite-float.zip) For more device-specific assets and performance metrics, visit **[Depth-Anything-V3 on Qualcomm® AI Hub](https://aihub.qualcomm.cn/models/depth_anything_v3)**. ### Option 2: Export with Custom Configurations Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/depth_anything_v3) Python library to compile and export the model with your own: - Custom weights (e.g., fine-tuned checkpoints) - Custom input shapes - Target device and runtime configurations This option is ideal if you need to customize the model beyond the default configuration provided here. See our repository for [Depth-Anything-V3 on GitHub](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/depth_anything_v3) for usage instructions. ## Model Details **Model Type:** Model_use_case.depth_estimation **Model Stats:** - Input resolution: 518x518 - Model checkpoint: da3-small - Model size (float): 94.3 MB - Number of parameters: 24.7M ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | Depth-Anything-V3 | ONNX | float | Snapdragon® X2 Elite | 28.251 ms | 5 - 5 MB | NPU | Depth-Anything-V3 | ONNX | float | Snapdragon® X Elite | 71.311 ms | 70 - 70 MB | NPU | Depth-Anything-V3 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 55.467 ms | 0 - 619 MB | NPU | Depth-Anything-V3 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 131.005 ms | 2 - 642 MB | NPU | Depth-Anything-V3 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 77.605 ms | 3 - 10 MB | NPU | Depth-Anything-V3 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 72.464 ms | 0 - 79 MB | NPU | Depth-Anything-V3 | ONNX | float | Qualcomm® QCS8450 | 131.005 ms | 2 - 642 MB | NPU | Depth-Anything-V3 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 78.308 ms | 3 - 9 MB | NPU | Depth-Anything-V3 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 71.311 ms | 70 - 70 MB | NPU | Depth-Anything-V3 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 36.375 ms | 2 - 482 MB | NPU | Depth-Anything-V3 | ONNX | float | Snapdragon® 8 Elite Mobile | 36.375 ms | 2 - 482 MB | NPU | Depth-Anything-V3 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 28.468 ms | 2 - 507 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Snapdragon® X2 Elite | 34.582 ms | 3 - 3 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Snapdragon® X Elite | 80.612 ms | 3 - 3 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 60.937 ms | 3 - 631 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 134.845 ms | 2 - 618 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 85.964 ms | 3 - 10 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 81.015 ms | 3 - 6 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® SA8775P | 86.658 ms | 1 - 462 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® SA8650P | 86.658 ms | 1 - 462 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® SA8255P | 86.658 ms | 1 - 462 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® QCS8450 | 134.845 ms | 2 - 618 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 86.12 ms | 3 - 9 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 80.612 ms | 3 - 3 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 45.792 ms | 3 - 493 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® SA7255P | 185.859 ms | 1 - 474 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Qualcomm® SA8295P | 112.225 ms | 0 - 460 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 45.792 ms | 3 - 493 MB | NPU | Depth-Anything-V3 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 33.012 ms | 3 - 509 MB | NPU | Depth-Anything-V3 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 56.174 ms | 1 - 586 MB | NPU | Depth-Anything-V3 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 117.697 ms | 1 - 584 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 79.852 ms | 1 - 82 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 75.879 ms | 0 - 11 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® SA8775P | 81.55 ms | 1 - 454 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® SA8650P | 81.55 ms | 1 - 454 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® SA8255P | 81.55 ms | 1 - 454 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® QCS8450 | 117.697 ms | 1 - 584 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 81.642 ms | 1 - 81 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 41.753 ms | 1 - 476 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® SA7255P | 181.716 ms | 2 - 455 MB | NPU | Depth-Anything-V3 | TFLITE | float | Qualcomm® SA8295P | 102.962 ms | 1 - 456 MB | NPU | Depth-Anything-V3 | TFLITE | float | Snapdragon® 8 Elite Mobile | 41.753 ms | 1 - 476 MB | NPU | Depth-Anything-V3 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 30.922 ms | 1 - 479 MB | NPU ## License * The license for the original implementation of Depth-Anything-V3 can be found [here](https://github.com/ByteDance-Seed/Depth-Anything-3/blob/main/LICENSE). ## References * [Depth Anything 3: Recovering the visual space from any views](https://arxiv.org/abs/2511.10647) * [Source Model Implementation](https://github.com/ByteDance-Seed/Depth-Anything-3) ## Community * For questions or feedback, chat with our engineers in [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.cn).