# Image Generation

Generate and edit images directly inside your chatbot using multiple AI image models.

---

Chipp apps can generate and edit images directly in the chat conversation. Your consumers simply describe what they want, and the AI creates it -- no separate tools or plugins required.

## Available Models

Chipp supports five image generation models from four providers. Each model has different strengths, costs, and capabilities.

| Model | Provider | Best For | Cost per Image | Max Resolution |
|-------|----------|----------|----------------|----------------|
| **Gemini 3 Pro** | Google | Highest quality, text rendering, multi-reference editing | ~$0.17 | 4K |
| **GPT Image 1.5** | OpenAI | Precise instruction-following edits | ~$0.04 | 4096x4096 |
| **FLUX.1 Kontext Pro** | Black Forest Labs | Conversational editing, preserving originals | ~$0.05 | 1024x1024 |
| **Stability AI SD 3.5** | Stability AI | Inpainting, search-and-replace, background removal | ~$0.05 | 1024x1024 |
| **Gemini 2.5 Flash** | Google | Fast, affordable, high-volume generation | ~$0.003 | Dynamic |

> **Note:** GPT Image 1.5 is the default model for new apps. Gemini 3 Pro is the recommended premium option for apps that need the highest quality output.

### Model Capabilities

Not every model supports every operation:

- **Generation** -- Create images from text descriptions. All models support this.
- **Editing** -- Modify an existing image based on instructions. Supported by all models except Gemini 2.5 Flash has limited editing.
- **Inpainting** -- Fill in or replace specific regions of an image. Supported by Gemini 3 Pro and Stability AI.
- **Blending** -- Combine multiple reference images into a new composition. Supported by Gemini 3 Pro and Gemini 2.5 Flash.

## Enabling Image Generation

Image generation is enabled by default for all apps on all tiers. To choose a model or adjust settings:


**Open Build Settings.** 
  Navigate to your app in the builder and open the **Build** tab.


**Find the Image Model Section.** 
  Scroll down to the **Image Model** card. You will see a grid of available models.


**Select a Model.** 
  Click on the model you want to use. The selected model will be highlighted with your brand color.


**Adjust Model Settings.** 
  Each model exposes different configuration options. Adjust them below the model grid after selecting a model.


**Save.** 
  Click **Save** at the top of the Build tab. The new model will be used for all future image generation requests.


## Model Settings

Each model offers different configuration parameters that your consumers do not see -- these are builder-level defaults applied to every generation.

### GPT Image 1.5 (OpenAI)

- **Quality** -- Low (fastest, cheapest), Medium (balanced, default), or High (best quality). Higher quality costs more tokens.
- **Size** -- 1024x1024 (square), 1536x1024 (landscape), or 1024x1536 (portrait).
- **Background** -- Auto (model decides), Transparent (PNG only, useful for logos and stickers), or Opaque (solid background).

### Gemini 3 Pro

- **Aspect Ratio** -- Choose from 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, or 2:3. The model adapts the composition to fit the chosen ratio.

### FLUX.1 Kontext Pro

- **Aspect Ratio** -- Same options as Gemini 3 Pro.
- **Prompt Upsampling** -- When enabled, the model automatically enhances your consumer's prompt for better results.

### Stability AI SD 3.5

- **Aspect Ratio** -- Similar options. Only applies to new generations; edits preserve the original dimensions.
- **Negative Prompt** -- Describe elements to exclude (e.g., "blurry, low quality, watermark"). Guides the model away from unwanted content.
- **Edit Strength** -- Slider from 0.1 to 1.0. Lower values make subtle tweaks; higher values allow dramatic transformations. Default is 0.7.

## How It Works for Consumers

Consumers interact with image generation through natural language. They do not need to know which model is configured -- they just ask.

### Generating New Images

Consumers describe what they want, and the AI calls the `generateImage` tool automatically:

- "Create a logo for my coffee shop called Bean There"
- "Generate a watercolor painting of a sunset over mountains"
- "Make me a product mockup of a blue water bottle"

The generated image appears inline in the chat conversation.

### Editing Existing Images

Consumers can upload an image and ask for changes. The AI uses the uploaded image as a reference:

- "Make the sky more orange" (after uploading a landscape photo)
- "Remove the background" (after uploading a product photo)
- "Add a hat to the person in this photo"

There is also a dedicated `editImage` tool that automatically resolves the most recently uploaded image, so consumers can simply say "change the color to blue" without re-uploading.

### Multi-Image Blending

With Gemini models, consumers can upload multiple reference images and ask the AI to combine them:

- "Blend these two photos into a single composition"
- "Use this character from image 1 in the scene from image 2"

> **Tip:** Gemini 3 Pro supports up to 14 reference images in a single request, making it the best choice for complex multi-image workflows.

## Disabling Image Generation

If your app does not need image generation, you can disable it in the Build tab. When disabled, the `generateImage` and `editImage` tools are removed from the agent's toolset entirely -- consumers will not be able to request images.

## Billing

Image generation is billed per image through your organization's Stripe Token Billing balance. The cost per image depends on the model you select (see the table above). Costs shown in the builder include a 30% platform markup over the raw provider cost.