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

# Image models

> Compare 20+ image models across Flux, Nano Banana, Seedream, GPT Image, and other families — costs range from 1 credit (Flux 2 Flash) to 21 credits (GPT Image) per generation at default settings.

All costs below are **indicative base credits** at default settings (1K resolution). Resolution multipliers and quality tiers apply on top of the base cost.

<Tip>
  Use **`auto`** in Studio or the MCP `generate_image` tool to let Elaris route automatically to a strong default model.
</Tip>

<Tabs>
  <Tab title="Flux">
    **Black Forest Labs** — fast, versatile, and cost-effective across all quality tiers.

    | Model              | Credits | Notes                         |
    | ------------------ | ------- | ----------------------------- |
    | Flux 2 Flash       | 1       | Ultra-fast; lowest cost       |
    | Flux 2 Turbo       | 1       | Fast with improved quality    |
    | Flux 2             | 2       | Balanced quality and speed    |
    | Flux 2 Flex        | 5       | Multi-reference input support |
    | Flux 1 Kontext Pro | 5       | Precise in-context editing    |
    | Flux 2 Pro         | 7       | Professional-grade output     |
    | Flux 2 Max         | 8       | Maximum quality               |
  </Tab>

  <Tab title="Nano Banana">
    **Google** — strong character consistency across generations; Nano Banana 2 is an auto-router candidate.

    | Model            | Credits | Notes                                        |
    | ---------------- | ------- | -------------------------------------------- |
    | Nano Banana      | 5       | Affordable baseline                          |
    | Nano Banana Lite | 7       | Fast edits                                   |
    | Nano Banana 2    | 10      | Character consistency; auto-router candidate |
    | Nano Banana Pro  | 19      | Premium quality                              |
  </Tab>

  <Tab title="Seedream">
    **ByteDance** — competitive quality at mid-range credit costs.

    | Model            | Credits | Notes                    |
    | ---------------- | ------- | ------------------------ |
    | Seedream 4       | 4       | Efficient baseline       |
    | Seedream 4.5     | 5       | Improved detail          |
    | Seedream v5 Lite | 5       | Latest efficient variant |
  </Tab>

  <Tab title="GPT Image">
    **OpenAI** — best-in-class text rendering within images; supports quality tier selection.

    | Model         | Credits | Notes                             |
    | ------------- | ------- | --------------------------------- |
    | GPT Image 1   | 21      | Strong text-in-image rendering    |
    | GPT Image 1.5 | 21      | Production-ready quality          |
    | GPT Image 2   | 21      | Fine-grained edits; quality tiers |

    <Note>
      GPT Image models support quality tier selection in Studio. Higher quality tiers may increase the credit cost above the base 21.
    </Note>
  </Tab>

  <Tab title="Other families">
    Additional model families for specialized generation and editing workflows.

    | Family               | Models                               | Best for                       |
    | -------------------- | ------------------------------------ | ------------------------------ |
    | **Kling** (Kuaishou) | Kling O1 Image                       | Multi-reference editing        |
    | **Wan** (Alibaba)    | Wan 2.5, 2.7, 2.7 Pro                | General and premium generation |
    | **Runway**           | Gen-4 Image, Gen-4 Turbo             | Cinematic, film-style output   |
    | **Ideogram**         | Replace BG, Reframe, Edit, Remove BG | Utility edits and cleanup      |
    | **Phota**            | Phota, Phota Edit, Phota Enhance     | Photorealistic portraiture     |
    | **xAI**              | Grok Imagine                         | Creative / experimental        |
    | **Tencent**          | Hunyuan v3                           | Strong text rendering          |
    | **QuiverAI**         | Arrow 1.0, 1.1, 1.1 Max              | SVG vector generation          |
  </Tab>
</Tabs>

<Note>
  The model catalog is updated regularly. Call the MCP tool `list_models` with `kind: "image"` to fetch the live list with current credit costs.
</Note>

<CardGroup cols={2}>
  <Card title="How pricing works" icon="calculator" href="/docs/pricing/how-pricing-works">
    Understand resolution multipliers and the full credit formula.
  </Card>

  <Card title="Video models" icon="video" href="/docs/models/video">
    Browse video generation families and per-second pricing.
  </Card>
</CardGroup>