> For the complete documentation index, see [llms.txt](https://docs.maiagent.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.maiagent.ai/maiagent-user-guide/en/build/tools/ai-image-generation-guide.md).

# Built-in AI Image Generation Tool

This guide introduces MaiAgent's built-in AI image generation tools and compares their results and use cases

### 📋 Feature Overview <a href="#feature-summary" id="feature-summary"></a>

MaiAgent includes three AI image generation tools that can create new images and edit images based on references. Select them in your Agent's tool settings to start generating images right away.

On the Agent settings page, click “<mark style="color:blue;">Select Tools</mark>” under “<mark style="color:blue;">Tools</mark>”, then select the following image generation tools under “<mark style="color:blue;">System Built-in Tools</mark>”:

* <mark style="color:blue;">Gemini 3 Pro Image</mark>
* <mark style="color:blue;">Gemini 3.1 Flash Image</mark>
* <mark style="color:blue;">GPT Image 2 Image Generation Tool</mark>

<figure><img src="https://1360999650-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6v6TNkkOQVfRYfcNirHL%2Fuploads%2Fgit-blob-79b7d7d056695fd0b39063bb5214a57698409eb5%2Fai-image-tools-select-tools.png?alt=media" alt="Gemini 3.1 Flash Image, GPT Image 2 Image Generation Tool, and Gemini 3 Pro Image in the Select Tools dialog"><figcaption><p>Image generation tools in the Select Tools dialog</p></figcaption></figure>

{% hint style="info" %}
See: [Configure Tools for an Agent](/maiagent-user-guide/en/build/tools/configure_tools.md)
{% endhint %}

***

### 🎨 Full Tool Comparison <a href="#tool-comparison-table" id="tool-comparison-table"></a>

| Tool                       | Key Features                                                                                  | Use Cases                                                                 |
| -------------------------- | --------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------- |
| **Gemini 3 Pro Image**     | Context understanding, Chinese prompts, reference image input, conversational editing         | Creative tasks requiring context understanding, Chinese-language requests |
| **Gemini 3.1 Flash Image** | Fast generation, Chinese prompts, reference image input, editing with multi-step instructions | Rapid iteration, creative brainstorming                                   |
| **GPT Image 2**            | Professional quality, iterative refinement, reference image input                             | Brand design, professional use                                            |

### ⭐️ Generation Examples <a href="#generation-examples" id="generation-examples"></a>

#### GPT Image 2 - Professional Design Example <a href="#gpt-image-professional-example" id="gpt-image-professional-example"></a>

**Previous generation example:** The coffee cup logo below was generated earlier and is retained to illustrate its line work and shape.

{% hint style="info" %}
This older image has a transparent background, which does not mean direct generation of transparent backgrounds is still supported. If you need an asset with its background removed, use the solid-color background approach under “How do I generate an image with a transparent background?” below.
{% endhint %}

<figure><img src="https://1360999650-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6v6TNkkOQVfRYfcNirHL%2Fuploads%2Fgit-blob-adda4bd85627865956b8d155649cd637e98c005f%2Fai-image-gpt-image-2-logo-example.png?alt=media" alt="Minimalist coffee cup logo previously generated with GPT Image 2, retaining the older image with a transparent background"><figcaption><p>Previous generation example: a minimalist coffee cup logo. The transparent background does not represent current generation capabilities.</p></figcaption></figure>

**Features:** Refined line work, suitable for branding

***

### 📝 Usage and Best Practices <a href="#usage-and-best-practices" id="usage-and-best-practices"></a>

#### Basic Usage Syntax <a href="#basic-syntax" id="basic-syntax"></a>

| Use Case                | Example Prompt                                                   | Recommended Tool       |
| ----------------------- | ---------------------------------------------------------------- | ---------------------- |
| **Everyday creations**  | `Draw a cute puppy`                                              | Gemini 3 Pro Image     |
| **Professional design** | `Design a modern, minimalist logo with a solid white background` | GPT Image 2            |
| **Rapid prototyping**   | `Quickly generate a concept image for a website homepage`        | Gemini 3.1 Flash Image |

#### Iterative Refinement (GPT Image 2) <a href="#multi-round-iteration-gpt-image" id="multi-round-iteration-gpt-image"></a>

```
Round 1: “Design a coffee shop logo”
Round 2: “Change the color to dark brown”
Round 3: “Add some steam effects”
Round 4: “Make the overall design more minimalist”

```

#### Editing with Reference Images <a href="#image-reference-editing-gemini-native" id="image-reference-editing-gemini-native"></a>

Gemini 3 Pro Image, Gemini 3.1 Flash Image, and GPT Image 2 all support editing with reference images. Upload the image you want to modify in the conversation and describe only the parts you want to change. The tools will try to keep anything you do not mention unchanged:

```
“Based on this image, change the background to a seaside scene”
“Keep the person unchanged and only change the clothing color”
“Add some flowers to this scene”

```

{% hint style="info" %}
If the Agent says it cannot read a previously generated image, upload the image you want to modify to the conversation again, then describe your requested changes.
{% endhint %}

***

### 🎯 Practical Use Case Guide <a href="#use-case-guide" id="use-case-guide"></a>

#### **Scenario 1: Social Media Content Creation** <a href="#use-case-social-media-content" id="use-case-social-media-content"></a>

**Need:** Create an image for an Instagram post

**Recommendation:** Gemini 3 Pro Image

**Example prompt:** `Create a cozy café scene suitable for an IG post`

#### **Scenario 2: Corporate Brand Design** <a href="#use-case-enterprise-brand-design" id="use-case-enterprise-brand-design"></a>

**Need:** Design a company logo and brand assets

**Recommendation:** GPT Image 2

**Example prompt:** `Design a logo for a technology company in a minimalist, modern style with a solid white background`

#### **Scenario 3: Creative Ideation and Prototyping** <a href="#use-case-creative-ideation-prototype" id="use-case-creative-ideation-prototype"></a>

**Need:** Quickly visualize creative concepts and refine them through iteration

**Recommendation:** Gemini 3.1 Flash Image

**Example prompt:** `Concept art for a mobile app interface design`

***

### ❓ Frequently Asked Questions and Solutions <a href="#faq-and-solutions" id="faq-and-solutions"></a>

#### Image Quality Questions <a href="#image-quality-issues" id="image-quality-issues"></a>

**Q: How do I get the highest-quality images?**

**A:** Use GPT Image 2 or Gemini 3 Pro Image and provide a detailed description:

* ✅ Specific style requirements (such as “professional photography style”)
* ✅ Detailed scene descriptions (lighting, angle, atmosphere)
* ✅ Clear quality requirements (such as “high resolution” or “commercial quality”)

**Q: Why doesn't the generated image match my expectations?**

**A:** Suggestions for improving your prompts:

* 🎯 Use specific rather than abstract descriptions
* 🎨 Specify a clear artistic style
* 📐 Explain composition and viewing angle requirements
* 🌈 Describe colors and lighting effects

#### Feature Usage Questions <a href="#feature-usage-issues" id="feature-usage-issues"></a>

**Q: How do I generate an image with a transparent background?**

**A:** Currently, none of the three image generation tools (Gemini 3 Pro Image, Gemini 3.1 Flash Image, and GPT Image 2) can directly generate images with transparent backgrounds. If you need an asset with its background removed (a product image, logo, or icon), specify a single solid-color background in your description so the subject is clearly separated from the background, then remove the background with a background removal tool:

```
Design a logo with a solid white background and no shadows between the subject and the background
Generate a product photo with the product on a solid green background

```

**Q: Can I modify a generated image?**

**A:** Yes! Gemini 3 Pro Image, Gemini 3.1 Flash Image, and GPT Image 2 all support editing with reference images. Upload the image you want to modify, then describe the parts you want to change:

```
Based on this image, change the sky to sunset colors
Keep the composition unchanged and only change the person's clothing

```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.maiagent.ai/maiagent-user-guide/en/build/tools/ai-image-generation-guide.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.