---
license: apache-2.0
library_name: rkllm
tags:
- rk182x
- rockchip
- rknpu
- llm
- qwen
- edge-ai
---

# Depth-Anything-V3 — [Depth]视觉深度信息v3

[English](README.md) | [简体中文](README_CN.md)

## 1. Overview

视觉深度信息v3，给定图像输出逐像素深度图与置信度（Depth Anything V3 单目深度估计）

This repository ships the RK182x deployment artifacts (RKNN graphs +
weights) **together with the wdnwebui worker manifest**
(`DepthAnythingV3.json`).

| Item | Value |
| --- | --- |
| Model ID | `wardner/Depth-Anything-V3` |
| Display name | Depth-Anything-V3 |
| Model type | LLM |
| Compute target | RK182x |
| Host platform | RK3588 |
| Runtime | RKLLM + RKNN |
| Capabilities | `swcap_vision_depth` |
| Worker factory | `depth_anything_v3` |
| Memory estimate | ~3000 MB |
| Init time | ~30 s |
| Cards required | 1 |
| Total weight size | 1.04 GB |

## 2. Files

| Path | Description | Size |
| --- | --- | --- |
| `global_DepthAnythingV3.280x280.1view.1828.rknn` | RKNN graph (RK182x compute card, 1.82 TOPS generation) | 2.7 MB |
| `global_DepthAnythingV3.280x280.1view.weight` | Weights paired with the RKNN graph above | 552.7 MB |
| `head_DepthAnythingV3.280x280.1view.1828.rknn` | RKNN graph (RK182x compute card, 1.82 TOPS generation) | 1.8 MB |
| `head_DepthAnythingV3.280x280.1view.weight` | Weights paired with the RKNN graph above | 188.0 MB |
| `local_DepthAnythingV3.280x280.1view.1828.rknn` | RKNN graph (RK182x compute card, 1.82 TOPS generation) | 1.7 MB |
| `local_DepthAnythingV3.280x280.1view.weight` | Weights paired with the RKNN graph above | 316.6 MB |
| `DepthAnythingV3.json` | **worker manifest** — deliberately plain text | 1943 B |

> `DepthAnythingV3.json` is **deliberately not** in Git LFS. It is the
> manual of this distribution unit and must stay plain text so it can be
> diffed, reviewed and previewed on the Hub.

## 3. Download

```bash
hf download wardner/Depth-Anything-V3 --local-dir ./Depth-Anything-V3
```

Or with git (requires Git LFS):

```bash
git clone https://huggingface.co/wardner/Depth-Anything-V3
```

If `huggingface.co` is unreachable from your network (DNS interception /
blocked access), point the client at a mirror instead:

```bash
HF_ENDPOINT=https://hf-mirror.com hf download wardner/Depth-Anything-V3 \
  --local-dir ./Depth-Anything-V3
```

Note: with `git clone` the weights are Git LFS **pointers** until
`git lfs pull` completes. `hf download` fetches the real files directly, and
does not create a `.git` directory — neither is required for the model to
work; see §4.

## 4. Installation (wdnwebui)

Clone the repository into the model root so the folder holds **both** the
weights and the worker manifest:

```bash
git clone https://huggingface.co/wardner/Depth-Anything-V3 /userdata/models/Depth-Anything-V3
```

Then open the web UI → **Compute** → the card appears under "local compute"
(the backend rescans the model root every ~2 s, **no restart needed**) →
click "enable".

The directory name does not have to match the model name: the SDK follows
"the folder the manifest was found in", located through the
`WDNAI_DEPTH_ANYTHING_V3_DIR` environment variable declared in the manifest.

## 5. SHA-256 verification

```bash
cd ./Depth-Anything-V3
sha256sum -c SHA256SUMS
```

All entries must report `OK` before deployment.

## 6. Caveats and limits

See the `notes` field of `DepthAnythingV3.json` — it records the
per-model specifics (sampling defaults, capability layout, side effects).

## 7. License

The upstream model is released under the Apache License 2.0. The files here
are RKNN-converted derivatives, not unmodified upstream weights. See
[LICENSE](LICENSE) and [NOTICE](NOTICE).