> ## Documentation Index
> Fetch the complete documentation index at: https://docs.videodraft.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Image Generation Models

> 20 AI models for image generation in AI Studio

## All Models

| Model            | Credits | Max Refs | Provider          | Best For                  |
| ---------------- | ------- | -------- | ----------------- | ------------------------- |
| VideoDraft Image | 10      | Multi    | VideoDraft        | Multi-image editing       |
| Imagen 4         | 4       | —        | Google            | Photorealism              |
| Imagen 4 Ultra   | 6       | —        | Google            | 4K quality                |
| Imagen 4 Fast    | 2       | —        | Google            | Fast previews             |
| Nano Banana      | 4       | 1        | Google            | Reference support         |
| Nano Banana Pro  | 14-24   | 14       | Google            | AI reasoning + 4K         |
| GPT-Image-1      | 8       | —        | OpenAI            | Creative concepts         |
| GPT-Image-1.5    | 1-20    | 16       | OpenAI            | Multi-input composition   |
| Flux 1.1 Pro     | 4       | —        | Black Forest Labs | Production quality        |
| Flux Dev         | 4       | —        | Black Forest Labs | Development testing       |
| Flux Kontext Pro | 4       | 1        | Black Forest Labs | Style consistency         |
| Flux Kontext Max | 8       | 1        | Black Forest Labs | Max consistency           |
| Flux 2 Pro       | 6       | 8        | Black Forest Labs | Character consistency     |
| Flux 2 Flex      | 12      | 10       | Black Forest Labs | Advanced controls         |
| Flux 2 Max       | 7+      | 8        | Black Forest Labs | Highest Flux quality      |
| Flux Schnell     | 1       | —        | Black Forest Labs | Quick drafts, iterations  |
| Flux Krea        | 3       | 1        | Black Forest Labs | Image-to-image            |
| Ideogram V3      | 3-8     | 1        | Ideogram          | Text rendering            |
| SeedDream 4      | 3       | 10       | ByteDance         | Multi-reference on budget |
| SeedDream 4.5    | 4       | 15       | ByteDance         | Multi-reference (beta)    |

<Info>
  Visionary, Studio, and Crew plans include unlimited image generation.
</Info>

## Multi-Reference Models

Use multiple input images for character/style consistency:

| Model           | Max Images | Credits     | Notes               |
| --------------- | ---------- | ----------- | ------------------- |
| Nano Banana Pro | 14         | 14-24       | AI reasoning        |
| GPT-Image-1.5   | 16         | 1-20        | Most inputs         |
| Flux 2 Pro      | 8          | 6           | Character projects  |
| Flux 2 Flex     | 10         | 12          | Advanced controls   |
| Flux 2 Max      | 8          | 7 + 3/image | Highest quality     |
| SeedDream 4     | 10         | 3           | Best budget option  |
| SeedDream 4.5   | 15         | 4           | Beta, 2K resolution |

## Variable Pricing Models

### GPT-Image-1.5

| Quality   | Square | Landscape/Portrait |
| --------- | ------ | ------------------ |
| Low       | 1      | 1                  |
| Medium    | 3      | 5                  |
| High/Auto | 13     | 20                 |

### Ideogram V3

| Mode     | Credits |
| -------- | ------- |
| Turbo    | 3       |
| Balanced | 6       |
| Quality  | 8       |

### Nano Banana Pro

| Resolution | Credits |
| ---------- | ------- |
| 1K / 2K    | 14      |
| 4K         | 24      |

## Resolution Support

| Model           | Max Resolution |
| --------------- | -------------- |
| Imagen 4        | 2048×2048      |
| Imagen 4 Ultra  | 4096×4096      |
| Nano Banana Pro | 4096×4096      |
| GPT-Image-1     | 1792×1024      |
| GPT-Image-1.5   | 2048×2048      |
| Flux series     | 1024×1024      |

## Recommended Workflow

<Steps>
  <Step title="Draft">
    Flux Schnell (1 credit) for rapid ideation
  </Step>

  <Step title="Refine">
    Flux 1.1 Pro or Imagen 4 (4 credits) for quality
  </Step>

  <Step title="Final">
    Imagen 4 Ultra or GPT-Image-1.5 for deliverables
  </Step>
</Steps>

## Aspect Ratios

All models support: 16:9 (landscape), 9:16 (portrait), 1:1 (square)

Most models also support: 4:3, 3:4, 3:2, 2:3

<Tip>
  Use Flux Schnell (1 credit) to test prompts before switching to premium models.
</Tip>