# Image Generation

BoltAI is not only great at generate text, it can also generate images using OpenAI's DALL·E or Stable Diffusion via Replicate.com

In this guide, you will learn how to effectively use BoltAI to generate AI images.

There are 2 ways you can generate images in BoltAI:

1. Use the text-to-image models directly.
2. Use the dedicated plugin.

### How to use DALL·E on macOS with BoltAI?

It's pretty straightforward. If you're using your own OpenAI API key:

1. Make sure you've set your OpenAI API Key.
2. Start a new chat, select OpenAI, choose Dall E 3 in the dropdown model
3. Enter your prompt in the chat input field.

If you're a Setapp user on an AI+ plan, you can simply choose **Setapp > Dall E 3** to start.

{% hint style="warning" %}
**NOTICE**

Unlike ChatGPT app, your request is sent directly to the Dall E model. Each request is completely different and BoltAI doesn't maintain a history between requests inside a chat.

For example, BoltAI would not understand when you say "Generate again".

If you prefer the ChatGPT workflow, consider using the Dall E plugin instead.
{% endhint %}

<figure><img src="/files/5LKZdnKuDmHrfhHl3Fvo" alt=""><figcaption></figcaption></figure>

### How to use the Dall E AI plugin?

To use the Dall E AI plugin, you will need to configure an OpenAI API key, and use an AI chat model that supports Function Calling capability.

Setapp users on an AI+ plan can use the "Dall E 3 (via Setapp)" plugin instead.

#### 1. Configure the Dall E plugin

Go to Settings > Plugins > Dall E 3. Choose tab "Settings". Enter your OpenAI API key and optionally set other Dall E configuration such as Quality, Size and Style.

#### 2. Using the Dall E plugin

Start a new chat, use a chat model that supports Function Calling (for example: OpenAI GPT-4o). Enable the Dall E plugin then enter your request

<figure><img src="/files/Jg9ans42j5XHTbU8csv2" alt=""><figcaption></figcaption></figure>

### What are the differences between using the plugin and using the text-to-image model directly?

<table><thead><tr><th width="229"></th><th width="283">Use text-to-image model directly</th><th>Use AI plugin</th></tr></thead><tbody><tr><td><strong>License Required</strong></td><td>Available on Plus license and above</td><td>Available on Premium license only</td></tr><tr><td><strong>Chat Context*</strong></td><td>No</td><td>Yes</td></tr><tr><td><strong>Can be used with a different LLM (Claude for example)</strong></td><td>No</td><td>Yes</td></tr><tr><td><strong>Original prompts are modified</strong></td><td>No*</td><td>Yes</td></tr></tbody></table>

**Notes:**

* **Chat Context**: when you use a text-to-image model directly, each request is independent, even if they are within the same chat. So when you say "generate again", the AI model would not understand.
* **Modified Prompts**: depends on the model you're using, the Chat model might modify your original prompt into something else. You can click the info button on each generated image to see the revised prompt.

### Can I use Stable Diffusion or another model?

Yes. You can use Stable Diffusion and other text-to-image models via Replicate. It's also available to use as an AI plugin, just like the Dall E plugin.

Replicate makes it easy to run machine learning models in the cloud from your own code. You can run open-source models, or deploy your own models.

1. Make sure you've set your Replicate.com API Key
2. Follow the same process to generate an image via Replicate

<figure><img src="/files/ALS4Usy9VGygLUArLhCq" alt=""><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

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

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.boltai.com/docs/chat-ui/image-generation.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.