[Integrations](https://spiral.ai/integrations) › [Gemini CLI](https://spiral.ai/integrations/gemini-cli) › [VOID Video Inpainting](https://spiral.ai/integrations/gemini-cli/void-video-inpainting)

# How to Use VOID Video Inpainting in Gemini CLI

VOID Video Inpainting removes objects from footage and rebuilds the physical interactions affected by their absence. Use it for edits where deleting a person or prop also needs to change what falls, collides, or keeps moving.

|  |  |
| --- | --- |
| Model | VOID Video Inpainting |
| Available through | fal |
| Category | Video |
| Price | Up to $0.195 per run at the default inputs. |
| Returns | video |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with VOID Video Inpainting in Gemini CLI

- Use VOID Video Inpainting on the short product clip in this project to remove the spare cup. Describe the finished scene as a product standing alone on the table, and save a comparison for review.
- Use VOID Video Inpainting on the project's obstacle clip to remove the block in the ball's path. Check the affected region along the ball's expected route and save the result separately.
- Use VOID Video Inpainting on the project's prop-handling clip to remove the performer. Review how the prop moves after losing support and flag frames where its shape changes.
- Use VOID Video Inpainting to compare the first pass with second-pass refinement on our existing removal test. Keep the source and mask the same, then show where object shape differs.
- Use VOID Video Inpainting on the project's 85-frame cleanup clip. Check the removal mask and affected region before rendering, then add the original and edited clips to the review page.

## How to set up VOID Video Inpainting in Gemini CLI

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

### Use Spiral's MCP server

1. **Add the server**: Or add it to `~/.gemini/settings.json` yourself:

```bash
gemini mcp add --scope user --transport http spiral https://api.spiral.ai/mcp
```

2. **Sign in**: Start `gemini` and run:

```text
/mcp auth spiral
```

3. **Check it**: In Gemini CLI, run `/mcp`. From your shell:

```bash
gemini mcp list
```

### Use the Spiral skill

1. **Send Gemini CLI 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 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 VOID Video Inpainting**: 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/void-video-inpainting",
    "input": {
      "video_url": "https://storage.googleapis.com/falserverless/kontext-blog/input_video.mp4",
      "prompt": "a video of buildings reflecting on a calm river"
    }
  }'
```

Then paste one of the prompts above: Gemini CLI finds VOID Video Inpainting on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/gemini-cli

## VOID Video Inpainting tools and prices

- [Void Video Inpainting](https://spiral.ai/tools/fal/fal-ai/void-video-inpainting) (fal): Up to $0.195 per run at the default inputs.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| quad_mask_video_url |  |  | URL of a mask video for the removal target. For best results this should be a VOID-style quadmask video with 4 grayscale values: 0=object to remove, 63=overlap, 127=affected region, 255=background to keep. A simple binary mask (0=remove, 255=keep) also works. If omitted, the app will generate a temporary mask video from `mask_prompt` using the SAM-3 video endpoint and convert it to a quadmask automatically. |
| mask_prompt |  | — | Text description of what should be masked in the input video, such as the object or person to remove. Used to generate a temporary mask video with SAM-3 when `quad_mask_video_url` is not provided. |
| enable_pass2_refinement |  | false | Run VOID Pass 2 warped-noise refinement after Pass 1. This is slower but can improve temporal consistency on longer clips. |
| negative_prompt |  | The video is not of a high quality, it has a low resolution. Watermark present in each frame. The background is solid. Strange body and strange trajectory. Distortion. | Negative prompt to guide generation away from undesired outputs. |
| num_inference_steps | 1 to 50 | 30 | Number of denoising steps. Higher values improve quality but increase latency. |
| guidance_scale | 0 to 20 | 1 | Classifier-free guidance scale. |

## What can you do with VOID Video Inpainting in Gemini CLI?

Turn project footage into removal drafts for an editor or review tool. A coding agent can organize source clips, masks, and comparisons while checking reconstructed motion and object shape.

### Clean up a filmed shot

Remove an unwanted prop from a short shot, then inspect its former location, shadow, and reflection before using the edit.

### Prepare an advertising edit

Create a draft with a distracting background object removed. Compare the reconstructed area with the original footage during review.

### Preview a scene without its performer

Explore how a shot could look after removing the person handling a prop, including any resulting change in the prop's motion.

### Explore a collision without an obstacle

Remove an obstacle from a filmed interaction and review the newly generated trajectory as a visual effects concept.

### Visualize loss of support

Remove a supporting object and generate a draft showing the supported object falling. Inspect the contact point and object shape throughout the motion.

### Build a cause-and-effect comparison

Pair an original interaction clip with an edited version that omits one object. Use the pair to illustrate a proposed alternative outcome for a lesson or presentation.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Gemini CLI guides](https://spiral.ai/integrations/gemini-cli)

## VOID Video Inpainting in other agents

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

## More Gemini CLI guides

- [VEED Video Background Removal Fast](https://spiral.ai/integrations/gemini-cli/veed-video-background-removal-fast)
- [Video Eraser Keypoints Video to Video](https://spiral.ai/integrations/gemini-cli/video-eraser-keypoints-video-to-video)
- [VEED Video Background Removal](https://spiral.ai/integrations/gemini-cli/veed-video-background-removal)
- [Video Eraser Mask Video to Video](https://spiral.ai/integrations/gemini-cli/video-eraser-mask-video-to-video)
- [VEED Subtitles](https://spiral.ai/integrations/gemini-cli/veed-subtitles)
- [Video Face Swap](https://spiral.ai/integrations/gemini-cli/video-face-swap)

## Questions about VOID Video Inpainting in Gemini CLI

**Can Gemini CLI use VOID Video Inpainting?**

Yes. Add Spiral to Gemini CLI as an MCP server, and Gemini CLI can find and run VOID Video Inpainting like any built-in tool.

**How much does VOID Video Inpainting cost in Gemini CLI?**

Up to $0.195 per run at the default inputs. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Gemini CLI plan is separate.

**Do I need a fal account to use VOID Video Inpainting in Gemini CLI?**

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

**How long does it take to set up VOID Video Inpainting in Gemini CLI?**

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

**Does Gemini CLI wait while VOID Video Inpainting runs?**

VOID Video Inpainting runs in the background. Gemini CLI starts the run, checks on it and gives you the video when it's ready.

**What is VOID Video Inpainting good at?**

Removing an object when its absence should change the action around it. That includes reconstructing falling objects, collisions, and motion paths, alongside effects such as shadows and reflections.

**Does VOID edit existing footage?**

Yes. VOID removes selected objects from an existing video and generates the scene that could remain without them.

**What should I write in the scene prompt?**

Describe the visible scene after the edit, such as "A ball rolls across the table." Keep the removal target separate from that description.

**When should I use the second refinement pass?**

Try it when a moving object changes shape in the first result. The second pass uses the generated motion to help stabilize object shape across frames.

**What resolution and length does the maker document?**

The published model card lists a default output of 672 pixels wide by 384 pixels high and a maximum of 197 frames. These are the maker's documented model specifications.

