[Integrations](https://spiral.ai/integrations) › [Hermes](https://spiral.ai/integrations/hermes) › [Depth Anything Video](https://spiral.ai/integrations/hermes/depth-anything-video)

# How to Use Depth Anything Video in Hermes

Depth Anything Video turns existing footage into depth maps that stay consistent between frames, useful for compositing and 3D conversion experiments. It estimates relative depth, so its values should not be treated as measured distances.

|  |  |
| --- | --- |
| Model | Depth Anything Video |
| Available through | fal |
| Category | Video |
| Price | Up to $3.12 per run at the default inputs for 60 seconds of input. |
| Returns | video |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Depth Anything Video in Hermes

- Each morning, use Depth Anything Video on newly submitted clips under 20 seconds. Prepare grayscale depth previews and a review list linking each to its source.
- Every Friday, process the week's selected landscape clips with Depth Anything Video at 720p. Organize the results for fog and haze experiments.
- Each evening, use Depth Anything Video on approved 10-second product clips. Prepare depth previews for background blur tests and list the shots awaiting review.
- Every Monday, prepare the selected short clips for 3D conversion tests with Depth Anything Video. Keep each depth result paired with its original video.
- Each day, use Depth Anything Video on new approved shots containing no more than 2,400 frames. Request raw depth arrays and prepare a processing summary.

## How to set up Depth Anything Video in Hermes

Pick the way that suits you. Each takes about two minutes.

### Use the Spiral skill

1. **Send Hermes this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use Spiral's MCP server

1. **Add the server and sign in**: Hermes opens your browser: sign in to Spiral and click Authorize. No API key. When Hermes asks Enable all tools? \[Y/n/select], answer Y: n cancels without saving the server, and select hides the tools you leave out.

```bash
hermes mcp add spiral \
  --url https://api.spiral.ai/mcp \
  --auth oauth \
  --connect-timeout 300
```

2. **Check it**: In a chat that was already open, run `/reload-mcp`.

```bash
hermes mcp list
hermes mcp test spiral
```

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run Depth Anything Video**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/depth-anything-video",
    "input": {
      "video_url": "https://v3b.fal.media/files/b/0a8fb1c1/xNTrr7wtczzLBkJdyE5_f_7JTYCmQe.mp4"
    }
  }'
```

Then paste one of the prompts above: Hermes finds Depth Anything Video on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/hermes

## Depth Anything Video tools and prices

- [Depth Anything Video](https://spiral.ai/tools/fal/fal-ai/depth-anything-video) (fal): `resolution`: 360p $0.052, 480p $0.052, 720p $0.052, auto (default) $0.052, 1080p $0.052 per second.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| model | VDA-Small, VDA-Base, VDA-Large | VDA-Large | Depth estimation model size. VDA-Large = best quality, VDA-Small = fastest. |
| colormap | grayscale, turbo, inferno, magma, viridis | grayscale | Colormap for depth visualization. 'turbo' (recommended) shows near=warm, far=cool. 'grayscale' for raw normalized depth. 'inferno'/'magma' for perceptually uniform. 'viridis' for colorblind-friendly. |
| resolution | auto, 360p, 480p, 720p, 1080p | auto | Output resolution. 'auto' preserves input (max 1080p). Options: 'auto', '360p', '480p', '720p', '1080p'. |
| max_frames |  | — | Max frames to process (max 2400). None = up to 2400. |
| output_fps |  | — | Output video FPS. None = same as input. |
| include_raw_depths |  | false | Export raw float32 depths as .npz file with: 'depths' [N,H,W], 'min_depth', 'max_depth', 'fps', 'model', 'shape'. |

## What can you do with Depth Anything Video in Hermes?

Prepare depth passes for newly submitted footage on a recurring schedule. Have the agent organize previews for review before anyone uses them in an edit.

### Depth passes for compositing

Generate a depth pass from an existing shot, then use it as a starting point for effects controlled by foreground and background distance.

### Background blur experiments

Use estimated depth to prototype background blur on filmed footage. Review subject boundaries and motion before accepting the composite.

### Fog and haze previews

Try stronger haze in distant parts of a shot using its depth pass. Compare the result with the source throughout the camera movement.

### Depth-based color treatments

Prototype different color treatments for near and distant regions. Use the depth estimate as a starting mask and refine the transition in your editor.

### 3D video conversion tests

Prepare depth estimates for a separate 2D-to-3D conversion workflow. Start with a short shot and review the converted sequence before processing more footage.

### Depth consistency reviews

Inspect depth predictions across a moving shot before using them in an effect. Compare the beginning, middle, and end, then watch the complete sequence.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Hermes guides](https://spiral.ai/integrations/hermes)

## Depth Anything Video in other agents

- [Amp](https://spiral.ai/integrations/amp/depth-anything-video)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/depth-anything-video)
- [Claude](https://spiral.ai/integrations/claude-desktop/depth-anything-video)
- [Claude Code](https://spiral.ai/integrations/claude-code/depth-anything-video)
- [Cline](https://spiral.ai/integrations/cline/depth-anything-video)
- [Codex](https://spiral.ai/integrations/codex/depth-anything-video)
- [Cursor](https://spiral.ai/integrations/cursor/depth-anything-video)
- [Devin](https://spiral.ai/integrations/devin/depth-anything-video)
- [dots](https://spiral.ai/integrations/dots/depth-anything-video)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/depth-anything-video)
- [Gemini](https://spiral.ai/integrations/gemini/depth-anything-video)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/depth-anything-video)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/depth-anything-video)
- [goose](https://spiral.ai/integrations/goose/depth-anything-video)
- [Grok](https://spiral.ai/integrations/grok/depth-anything-video)
- [Grok Build](https://spiral.ai/integrations/grok-build/depth-anything-video)
- [Instinct](https://spiral.ai/integrations/instinct/depth-anything-video)
- [Manus](https://spiral.ai/integrations/manus/depth-anything-video)
- [Muse](https://spiral.ai/integrations/muse/depth-anything-video)
- [Muse Code](https://spiral.ai/integrations/muse-code/depth-anything-video)
- [n8n](https://spiral.ai/integrations/n8n/depth-anything-video)
- [Notion](https://spiral.ai/integrations/notion/depth-anything-video)
- [OpenClaw](https://spiral.ai/integrations/openclaw/depth-anything-video)
- [opencode](https://spiral.ai/integrations/opencode/depth-anything-video)
- [Perplexity](https://spiral.ai/integrations/perplexity/depth-anything-video)
- [Raycast](https://spiral.ai/integrations/raycast/depth-anything-video)
- [VS Code](https://spiral.ai/integrations/vscode/depth-anything-video)
- [Windsurf](https://spiral.ai/integrations/windsurf/depth-anything-video)
- [Zapier](https://spiral.ai/integrations/zapier/depth-anything-video)

## More Hermes guides

- [DWPose Video](https://spiral.ai/integrations/hermes/dwpose-video)
- [DreamActor V2](https://spiral.ai/integrations/hermes/dreamactor-v2)
- [CassetteAI Video Sound Effects Generator](https://spiral.ai/integrations/hermes/cassetteai-video-sound-effects-generator)
- [ElevenLabs Dubbing](https://spiral.ai/integrations/hermes/elevenlabs-dubbing)
- [ByteDance Video Upscaler](https://spiral.ai/integrations/hermes/bytedance-video-upscaler)
- [FFmpeg](https://spiral.ai/integrations/hermes/ffmpeg)

## Questions about Depth Anything Video in Hermes

**Can Hermes use Depth Anything Video?**

Yes. Add Spiral to Hermes as an MCP server, and Hermes can find and run Depth Anything Video like any built-in tool.

**How much does Depth Anything Video cost in Hermes?**

Up to $3.12 per run at the default inputs for 60 seconds of input. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Hermes plan is separate.

**Do I need a fal account to use Depth Anything Video in Hermes?**

No. Spiral runs Depth Anything Video on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Depth Anything Video in Hermes?**

About two minutes: add the server and sign in, then check it.

**Does Hermes wait while Depth Anything Video runs?**

Depth Anything Video runs in the background. Hermes starts the run, checks on it and gives you the video when it's ready.

**What is Depth Anything Video useful for?**

It produces depth estimates from existing footage, with an emphasis on consistency between frames. Use those estimates as starting material for depth effects, compositing, or 3D video conversion.

**Does Depth Anything Video measure distances in meters?**

Treat its relative depth estimates as scene structure rather than measured distances. The original model shares a scale and shift across the video, but that does not establish an absolute distance scale.

**How much footage can Depth Anything Video process?**

The served endpoint caps processing at 2,400 frames: 80 seconds at 30 frames per second or 100 seconds at 24 frames per second. Split longer footage into shots for processing and review.

**What is the maximum output resolution?**

The served endpoint supports output up to 1080p. Preserving the input resolution automatically still applies that ceiling.

**Should I use a grayscale preview or raw depth arrays?**

Use a grayscale preview for visual inspection. Request raw depth arrays when your next step needs numerical processing, and keep the original footage alongside them for comparison.

