[Integrations](https://spiral.ai/integrations) › [Codex](https://spiral.ai/integrations/codex) › [Seedream 4.0 Edit](https://spiral.ai/integrations/codex/seedream-4-0-edit)

# How to Use Seedream 4.0 Edit in Codex

Use Seedream 4.0 Edit to change an existing image or carry a reference character, product, or style into a new scene. Choose it over Text to Image when the result needs to follow supplied visuals.

|  |  |
| --- | --- |
| Model | Seedream 4.0 Edit |
| Available through | fal |
| Category | Image |
| Price | $0.039 per image. Billed on `num_images`, default 1. |
| Returns | image |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Seedream 4.0 Edit in Codex

- Use Seedream 4.0 Edit with the product photograph in public/images/product.jpg. Replace the background with a sunlit workshop and keep the product centered. Request a 2048 × 2048 image.
- Use Seedream 4.0 Edit with design/reference-character.png and design/reference-scene.png. Place the character from the first image in the setting from the second. Request one 2048 × 2048 composition with light coming from the left.
- Remove the background from the photo I attach with Seedream 4.0 Edit and give me a transparent PNG. Save the result in ./assets.
- Use Seedream 4.0 Edit to clean up these product shots so they sit on pure white.
- Edit this image with Seedream 4.0 Edit: keep the subject, swap the background for a sunlit studio.

## How to set up Seedream 4.0 Edit in Codex

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

### Use Spiral's MCP server

1. **Add the hosted server**

```bash
codex mcp add spiral --url https://api.spiral.ai/mcp
```

2. **Sign in**: Codex opens the browser: sign in to Spiral and click Authorize. No API key. `codex mcp list` then shows Spiral.

```bash
codex mcp login spiral
```

### Use the Spiral skill

1. **Send Codex 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 Seedream 4.0 Edit**: 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/bytedance/seedream/v4/edit",
    "input": {
      "prompt": "Dress the model in the clothes and hat. Add a cat to the scene and change the background to a Victorian era building.",
      "image_urls": [
        "<image_urls>"
      ]
    }
  }'
```

Then paste one of the prompts above: Codex finds Seedream 4.0 Edit on Spiral, shows the price and runs it.

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

## Seedream 4.0 Edit tools and prices

- [Bytedance Seedream V4 Edit · Seedream 4.0 Edit](https://spiral.ai/tools/fal/fal-ai/bytedance/seedream/v4/edit) (fal): $0.039 per image. Billed on `num_images`, default 1.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| image_size |  | {"width":2048,"height":2048} | The size of the generated image. The minimum total image area is 921600 pixels. Failing this, the image size will be adjusted to by scaling it up, while maintaining the aspect ratio. |
| num_images | 1 to 6 | 1 | Number of separate model generations to be run with the prompt. |
| max_images | 1 to 6 | 1 | If set to a number greater than one, enables multi-image generation. The model will potentially return up to `max_images` images every generation, and in total, `num_images` generations will be carried out. In total, the number of images generated will be between `num_images` and `max_images*num_images`. The total number of images (image inputs + image outputs) must not exceed 15 |
| seed |  | — | Random seed to control the stochasticity of image generation. |
| enhance_prompt_mode | standard, fast | standard | The mode to use for enhancing prompt enhancement. Standard mode provides higher quality results but takes longer to generate. Fast mode provides average quality results but takes less time to generate. |

## What can you do with Seedream 4.0 Edit in Codex?

Create visual assets from project briefs or revise images already in the repository. Use the agent to turn written requirements into specific compositions and coordinated image sets.

### Storyboard a short scene

Request four related images with one character and one visual style. Describe each scene separately: arrival, discovery, decision, and departure.

### Develop a poster concept

Specify the poster's purpose, subject, layout, and exact headline. Use a 2048 × 2048 square for a first composition or request a larger image when the design needs more space.

### Revise an existing visual

Use Edit with a supplied image and a concrete instruction, such as replacing a background or changing an object's color. Name the target object directly.

### Create a coordinated illustration set

Describe a shared palette, medium, and recurring subject, then request a small series. Give each image a distinct scene while repeating the visual requirements.

### Combine visual references

Supply references for a character, setting, or style and explain the intended result. State which visual elements each reference contributes.

### Explore a new composition from text

Start with a written brief for an illustration or poster. Specify the subject's position, the environment, and the direction of light.

## Overview

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

## Seedream 4.0 Edit in other agents

- [Amp](https://spiral.ai/integrations/amp/seedream-4-0-edit)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/seedream-4-0-edit)
- [Claude](https://spiral.ai/integrations/claude-desktop/seedream-4-0-edit)
- [Claude Code](https://spiral.ai/integrations/claude-code/seedream-4-0-edit)
- [Cline](https://spiral.ai/integrations/cline/seedream-4-0-edit)
- [Cursor](https://spiral.ai/integrations/cursor/seedream-4-0-edit)
- [Devin](https://spiral.ai/integrations/devin/seedream-4-0-edit)
- [dots](https://spiral.ai/integrations/dots/seedream-4-0-edit)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/seedream-4-0-edit)
- [Gemini](https://spiral.ai/integrations/gemini/seedream-4-0-edit)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/seedream-4-0-edit)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/seedream-4-0-edit)
- [goose](https://spiral.ai/integrations/goose/seedream-4-0-edit)
- [Grok](https://spiral.ai/integrations/grok/seedream-4-0-edit)
- [Grok Build](https://spiral.ai/integrations/grok-build/seedream-4-0-edit)
- [Hermes](https://spiral.ai/integrations/hermes/seedream-4-0-edit)
- [Instinct](https://spiral.ai/integrations/instinct/seedream-4-0-edit)
- [Manus](https://spiral.ai/integrations/manus/seedream-4-0-edit)
- [Muse](https://spiral.ai/integrations/muse/seedream-4-0-edit)
- [Muse Code](https://spiral.ai/integrations/muse-code/seedream-4-0-edit)
- [n8n](https://spiral.ai/integrations/n8n/seedream-4-0-edit)
- [Notion](https://spiral.ai/integrations/notion/seedream-4-0-edit)
- [OpenClaw](https://spiral.ai/integrations/openclaw/seedream-4-0-edit)
- [opencode](https://spiral.ai/integrations/opencode/seedream-4-0-edit)
- [Perplexity](https://spiral.ai/integrations/perplexity/seedream-4-0-edit)
- [Raycast](https://spiral.ai/integrations/raycast/seedream-4-0-edit)
- [VS Code](https://spiral.ai/integrations/vscode/seedream-4-0-edit)
- [Windsurf](https://spiral.ai/integrations/windsurf/seedream-4-0-edit)
- [Zapier](https://spiral.ai/integrations/zapier/seedream-4-0-edit)

## More Codex guides

- [Sana Sprint](https://spiral.ai/integrations/codex/sana-sprint)
- [Seedream 4.0 Text to Image](https://spiral.ai/integrations/codex/seedream-4-0-text-to-image)
- [Sana 1.5 4.8B](https://spiral.ai/integrations/codex/sana-1-5-4-8b)
- [Seedream 4.5 Edit](https://spiral.ai/integrations/codex/seedream-4-5-edit)
- [Sana 1.5 1.6B](https://spiral.ai/integrations/codex/sana-1-5-1-6b)
- [Seedream 4.5 Text to Image](https://spiral.ai/integrations/codex/seedream-4-5-text-to-image)

## Questions about Seedream 4.0 Edit in Codex

**Can Codex use Seedream 4.0 Edit?**

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Seedream 4.0 Edit like any built-in tool.

**How much does Seedream 4.0 Edit cost in Codex?**

$0.039 per image. Billed on `num_images`, default 1. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

**Do I need a fal account to use Seedream 4.0 Edit in Codex?**

No. Spiral runs Seedream 4.0 Edit on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Seedream 4.0 Edit in Codex?**

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

**Does Codex wait while Seedream 4.0 Edit runs?**

Seedream 4.0 Edit runs in the background. Codex starts the run, checks on it and gives you the image when it's ready.

**When should I choose Edit instead of Text to Image?**

Choose Edit when you have a source image or visual references to follow. Choose Text to Image when your starting point is a written description; both use Seedream 4.0's shared generation and editing architecture.

**How large can Seedream 4.0 images be?**

The model supports up to 4K, including a 4096 × 4096 square. Custom dimensions must satisfy both the total pixel limit and the aspect ratio limit.

**Can it create a matching series of images?**

Yes. Related image sequences are a documented use for storyboards, comics, and coordinated designs. Describe the shared character and style, then specify what changes in each image.

**How should I request text inside an image?**

Put the exact wording in double quotation marks and state its role, such as the poster title or product label. For example, request a poster with the title "Weekend pottery".

**Do longer prompts produce better results?**

Very long prompts can cause the model to overlook details. BytePlus recommends staying under 600 English words; start with the subject, action, and setting, then add the visual requirements that matter.

