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

# Quick Start

> Create your first image or video, explore prompts, and connect through the API or MCP.

Choose where to start — browse the gallery, generate your own, or integrate via API and MCP.

<CardGroup cols={2}>
  <Card title="Gallery" icon="images" href="/en/features/gallery">
    Browse curated AI artwork from the community. Search, filter, and use prompts as starting points.
  </Card>

  <Card title="Image & Video Generation" icon="sparkles" href="/en/features/generate">
    Create images and videos with multiple AI models. Reference images, prompt enhancement, and advanced options.
  </Card>

  <Card title="MCP Plugin" icon="puzzle-piece" href="/en/mcp/overview">
    Generate images and videos directly from Claude Code, Cursor, Codex, Windsurf, Hermes Agent, OpenClaw, and any MCP host.
  </Card>

  <Card title="API Reference" icon="code" href="/en/api-reference/introduction">
    Integrate MeiGen into your applications with our REST API. Token-based authentication.
  </Card>
</CardGroup>

***

## Browse the Gallery

The gallery features curated, community-sourced AI artwork, refreshed daily.

**Filter by category** — use the sidebar tags to narrow down what you see:

* **Ads & Product** — product photography, packaging, marketing visuals
* **Brand & Logo** — logos, brand identity, visual systems
* **Videos** — hand-picked AI videos from top creators
* **Illustration & 3D** — digital art, concept art, 3D renders, character design
* **Posters & Visuals** — posters, key visuals, graphic and layout design
* **Portraits** — people, portraits, and photographic styles
* **Storyboard & Characters** — storyboard sequences and reusable character sheets
* **Wallpaper** — phone and desktop wallpapers

**Interact with cards** — hover over any image to reveal actions:

* **Use Idea** — copies the original prompt into the generation editor, ready to customize and generate
* **Save** — saves the image to your **Favorites**, synced across devices
* **View on X** — opens the original post on X (Twitter)

**Search** — on desktop, press `Cmd+K` (Mac) or `Ctrl+K` (Windows) to open the search dialog; **sign-in is required**. On mobile, the search page is open to everyone, no sign-in needed. Type a query and press `Enter` to search the gallery — hybrid search matches both the wording and the visual content of community prompts, and you can also search by uploading an image. Free accounts have a daily search allowance whether or not you're signed in; paid accounts are unlimited.

<Card title="Full Gallery Guide" icon="arrow-right" href="/en/features/gallery">
  Learn about image details, drag-to-reference, search, favorites, and more
</Card>

***

## Generate Images

The fastest start: find a gallery image you like and click **Use Idea** to copy it into the editor.

**Edit variables** — some prompts contain blue `[placeholder]` tags like `[style]` or `[subject]`. Click any tag to replace it with your own text. These are editable template variables that make it easy to create variations.

**Add reference images** — upload your own images or drag any gallery card directly into the reference area. When using multiple references, type `@` in the prompt editor to mention specific images (e.g., "Use @image1 as the background and @image2 for the character style").

**Enhance your prompt** — use the **Enhance Prompt** button to automatically improve short or simple prompts. The AI adds visual details like composition, lighting, colors, and textures.

**Choose your model** — select from a range of image models (including GPT Image 2.5 when available) and video models, each with different strengths. Seedance 2.5 supports 4–30 second videos at 480p, 720p or 1080p with generated audio — the soundtrack the model creates for you. That is separate from **reference audio**, your own WAV or MP3 clips attached as input to steer the result on Seedance 2.5 and 2.0; reference audio never costs credits. Pick an aspect ratio, resolution, and duration; the exact credit cost appears on the generate button before you submit.

<Card title="Full Generation Guide" icon="arrow-right" href="/en/features/generate">
  Learn about model selection, video generation, advanced options, and more
</Card>

***

GPT Image 2.5 offers five quality levels and a choice of Sunburst (quality priority) or Flare (speed priority), at the same price. Check [List Models](/en/api-reference/endpoint/models) for availability and the current default; see [model pricing](/en/features/models) for credit costs.

## Free Credits

<Tip>
  Every new account gets a welcome credit bonus, plus free daily credits that reset at 00:00 UTC. Daily credits cover the **Basic** models — currently **Flux 2 Klein** and **Agnes Image 2.5 Flash** for images; no video model is on the basic tier at the moment (Agnes Video 2.5 Flash has been temporarily unavailable since 2026-09-22). Premium models like GPT Image 2.5 and GPT Image 2, Nanobanana Pro, Seedream, Midjourney, Grok Imagine, Seedance, and Veo require welcome/purchased credits.
</Tip>

***

## Next Steps

<CardGroup cols={2}>
  <Card title="Generate Images & Videos" icon="wand-magic-sparkles" href="/en/features/generate">
    Prompt editor, references, video extension, enhancement, describe image
  </Card>

  <Card title="Model Comparison" icon="scale-balanced" href="/en/features/models">
    Compare available models — features, pricing, and limitations
  </Card>

  <Card title="MCP Plugin Setup" icon="plug" href="/en/mcp/setup">
    Install the MeiGen MCP plugin for Claude Code, Cursor, Codex, Windsurf, Hermes Agent, or OpenClaw
  </Card>

  <Card title="API Reference" icon="code" href="/en/api-reference/introduction">
    REST API documentation for programmatic access
  </Card>
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.