[Integrations](https://spiral.ai/integrations) › [Claude](https://spiral.ai/integrations/claude-desktop) › [NAFNet Denoise](https://spiral.ai/integrations/claude-desktop/nafnet-denoise)

# How to Use NAFNet Denoise in Claude

NAFNet Denoise reduces noise in existing photos, with research centered on smartphone images. Unlike NAFNet Deblur, it targets image noise rather than motion blur.

|  |  |
| --- | --- |
| Model | NAFNet Denoise |
| Available through | fal |
| Category | Image |
| Price | Up to $0.02925 per run at the default inputs. |
| Returns | image |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with NAFNet Denoise in Claude

- Use NAFNet Denoise on this attached smartphone photo. Show the original and result side by side, with matching crops of a smooth area and a detailed area.
- Use NAFNet Denoise on this product photo. Help me compare the packaging text and surface texture before I choose a version for my listing.
- Use NAFNet Denoise on this noisy room photo. Show enlarged comparisons of the wall, sofa fabric, and furniture edges.
- Use NAFNet Denoise on this indoor event photo. Compare faces, hair, and clothing details with the original.
- Use NAFNet Denoise on this dusk travel photo. Show the whole result and matching close-ups of the sky and distant buildings.

## How to set up NAFNet Denoise in Claude

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

### Use the Claude connector

1. **Add Spiral to Claude**: Click Add Spiral in Claude above. Claude opens Add custom connector with Spiral and its URL filled in; check them and click Continue. Without the link: in Claude open Customize › Connectors › Add › Add custom connector and paste this URL:

```text
https://api.spiral.ai/mcp
```

2. **Keep the detected sign-in and click Add**: Claude finds Spiral's sign-in by itself: leave Sign in now and Use Claude's published identity selected.

3. **Click Connect**: The connector page opens. Connect takes you to Spiral in your browser.

### Use the Spiral skill

1. **Send Claude 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 NAFNet Denoise**: 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/nafnet/denoise",
    "input": {
      "image_url": "https://storage.googleapis.com/falserverless/nafnet/noisy.png"
    }
  }'
```

Then paste one of the prompts above: Claude finds NAFNet Denoise on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/claude-desktop

## NAFNet Denoise tools and prices

- [NAFNet-denoise · Image to Image](https://spiral.ai/tools/fal/fal-ai/nafnet/denoise) (fal): Up to $0.02925 per run at the default inputs.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — | seed to be used for generation |

## What can you do with NAFNet Denoise in Claude?

Create a denoised candidate from a noisy photo you share in chat. Ask for matching detail crops so you can judge the result before keeping it.

### Smartphone photo cleanup

Denoise a noisy phone photo and compare enlarged crops before choosing the version to keep.

### Product photo review

Try noise reduction on product photos taken with a phone. Inspect packaging text, surface texture, and edges before using the result in a listing.

### Interior listing photos

Create denoised candidates from noisy room photos. Compare wall areas, furniture edges, and fabric details against the originals.

### Event photo selection

Test denoising on noisy indoor event photos. Review faces, hair, and clothing details before adding results to an album.

### Travel photo preparation

Try denoising phone photos taken at dusk. Review sky areas and distant details before selecting images for a travel story.

### Website photo preparation

Prepare denoised candidates for noisy photographic assets. Compare them at the intended display size and inspect enlarged details before replacing an original.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Claude guides](https://spiral.ai/integrations/claude-desktop)

## NAFNet Denoise in other agents

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

## More Claude guides

- [NAFNet Deblur](https://spiral.ai/integrations/claude-desktop/nafnet-deblur)
- [Nano Banana](https://spiral.ai/integrations/claude-desktop/nano-banana)
- [Muse Image Text to Image](https://spiral.ai/integrations/claude-desktop/muse-image-text-to-image)
- [Nano Banana 2](https://spiral.ai/integrations/claude-desktop/nano-banana-2)
- [Muse Image Edit](https://spiral.ai/integrations/claude-desktop/muse-image-edit)
- [Nano Banana 2 Image Editing](https://spiral.ai/integrations/claude-desktop/nano-banana-2-image-editing)

## Questions about NAFNet Denoise in Claude

**Can Claude use NAFNet Denoise?**

Yes. Add Spiral to Claude as an MCP server, and Claude can find and run NAFNet Denoise like any built-in tool.

**How much does NAFNet Denoise cost in Claude?**

Up to $0.02925 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 Claude plan is separate.

**Do I need a fal account to use NAFNet Denoise in Claude?**

No. Spiral runs NAFNet Denoise on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up NAFNet Denoise in Claude?**

About two minutes: add Spiral to Claude, then keep the detected sign-in and click Add, then click Connect.

**Does Claude wait while NAFNet Denoise runs?**

NAFNet Denoise runs in the background. Claude starts the run, checks on it and gives you the image when it's ready.

**What photos are a good starting point for NAFNet Denoise?**

Start with noisy smartphone photos. The published RGB denoising evaluation uses SIDD, a dataset of real smartphone images captured under different lighting conditions.

**Can NAFNet Denoise fix motion blur?**

Use it for noise reduction. The official implementation uses a separate deblurring model, so deblurring results in the NAFNet paper do not establish what NAFNet Denoise can do.

**Does NAFNet Denoise upscale an image?**

The published denoising architecture returns the original image height and width. Treat it as restoration at the existing dimensions; super-resolution uses a separate model.

**How should I prompt an agent to use NAFNet Denoise?**

Identify the photo to denoise and ask for an original-versus-result comparison. The published network processes image pixels without text conditioning, so style descriptions do not steer its restoration.

**How should I judge the result?**

Compare the original and result at matching magnification, then zoom into fine details. Check whether the noise reduction is useful without accepting unwanted changes to texture or edges.

