Image Generator Node
The Image Generator node is the primary image creation engine in Flow Studio. You feed it a prompt and optional reference assets, choose a model and style, and it produces high-quality AI-generated images at up to 4K resolution.
What It Does
This node takes text prompts (and optionally reference images) as input and generates images using AI models. You control the artistic style, aspect ratio, output resolution, and camera angle. When connected to a Prompt Generator node in Multiply mode, it can produce a batch of images -- one per prompt variation.
Configuration
| Parameter | Description | Options / Range |
|---|---|---|
| Model | The AI model used for generation | See models table below |
| Style | Artistic direction applied to the output | auto, realistic, artistic, anime, 3d |
| Aspect Ratio | Width-to-height ratio of the generated image | 1:1, 16:9, 9:16, 4:3, 3:4 |
| Size | Output resolution tier | 1K, 2K, 4K |
| Camera Angle | Perspective control for the generated scene | Configurable per generation |
| Multi-Angle Mode | Generates the same scene from multiple camera angles in one execution | On / Off |
Available Models
| Model | Credits/Image | Notes |
|---|---|---|
| Z-Image | 5 | Fastest. Text-to-image only, custom resolution (512-2048px) |
| Grok Imagine | 22 | Creative/expressive style (xAI via fal.ai) |
| Nano Banana (Gemini 2.5 Flash) | 40 | Fast, creative generation |
| SeeDream 4.5 | 40 | High quality, auto-ratio, custom resolution, 2K+ |
| Recraft V3 | 80 | Vector illustrations & icons |
| Nano Banana Pro (Gemini 3 Pro) 1K/2K | 150 | High quality, up to 14 reference images |
| GPT Image 2 | 10–410 | OpenAI. Quality tiers (low / medium / high), up to 16 reference images, best-in-class text rendering (Latin + CJK) |
| Nano Banana Pro (Gemini 3 Pro) 4K | 280 | Maximum quality 4K output |
GPT Image 2
GPT Image 2 introduces per-output quality tiers that directly affect cost and detail level. The quality selector defaults to high.
| Quality | Use Case | Credits Range |
|---|---|---|
| Low | Drafts, quick iterations | 10–60 cr |
| Medium | Standard production | 50–200 cr |
| High (default) | Final assets, hero images | 220–410 cr |
The exact cost depends on the chosen size. Larger sizes and higher quality tiers cost more.
Aspect ratios supported: 1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4, plus custom resolution (max edge 3840px, between 655K and 8.3MP, dimensions multiples of 16, aspect ratio ≤ 3:1).
Reference images: Up to 16 input images for character consistency, style transfer, or composition.
Text rendering: GPT Image 2 produces sharper, more legible text in Latin and CJK scripts than other models — a strong choice for posters, packaging, and infographics.
Edit mode and aspect ratio: When you supply a reference image to GPT Image 2, the model respects the aspect ratio you've configured on the node — it does not auto-size to the source image. Set the ratio explicitly (or use one of the custom resolutions above) to control the output dimensions.
Batch quality editing: Select multiple GPT Image 2 nodes on the canvas and the floating toolbar exposes a quality selector. Switching it updates the quality tier (low / medium / high) for every selected node at once.
Available Styles
| Style | Description |
|---|---|
auto | The model automatically selects the best style for the prompt |
realistic | Photorealistic rendering with natural lighting and detail |
artistic | Painterly, illustrative, or stylized output |
anime | Japanese animation aesthetic |
3d | Three-dimensional rendered appearance |
Aspect Ratios
| Ratio | Common Use Case |
|---|---|
1:1 | Social media posts, profile images, product shots |
16:9 | Landscape banners, video thumbnails, desktop wallpapers |
9:16 | Vertical stories, mobile wallpapers, portrait banners |
4:3 | Presentations, traditional photo prints |
3:4 | Portrait-oriented prints, book covers |
Usage
- Drag an Image Generator node onto the canvas.
- Connect a Prompt Generator to the prompt input and optionally connect an Asset Node as a reference image.
- Select your preferred Model, Style, Aspect Ratio, and Size.
- Optionally configure Camera Angle or enable Multi-Angle Mode.
- Execute the node to generate your image(s).
For product photography workflows, use realistic style with a 4:3 or 1:1 ratio and connect a product asset as a reference to maintain visual consistency.
Related Nodes
- Prompt Generator Node -- supplies the text prompt that drives generation.
- Asset Node -- provides reference images to guide the output.
- Video Generator Node -- animate a generated image into a video.
- Image Expand Node -- extend the boundaries of a generated image.
- Image Edit Node -- convert the output format.
- Remove Background Node -- isolate the subject from the background.