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

# GlobalAI OPC: Text, Image, and Video Generation APIs

> GlobalAI OPC provides video, image, and text AI generation APIs through a unified OpenAI-compatible gateway supporting Sora, VEO, Claude, GPT, and more.

GlobalAI OPC is a unified AI API gateway that lets you generate text, images, and videos through a single, consistent interface. Every endpoint is OpenAI-compatible, so you can integrate familiar tooling and SDKs while accessing a broad catalog of cutting-edge models — from OpenAI's Sora and GPT to Google's VEO and Gemini, Anthropic's Claude, and a growing roster of specialized video and image generators.

## Core API Categories

<CardGroup cols={2}>
  <Card title="Text APIs" icon="message" href="/api-reference/text/openai-chat">
    Generate and stream conversational text using GPT, Claude, and Gemini models through a single OpenAI-compatible chat completions endpoint.
  </Card>

  <Card title="Image APIs" icon="image" href="/api-reference/image/nano-banana-create">
    Create high-quality images from text prompts using Nano Banana and GPT Image models, with flexible aspect ratios and resolutions up to 4K.
  </Card>

  <Card title="Video APIs" icon="video" href="/api-reference/video/sora-create">
    Produce AI-generated videos from text or image prompts using Sora, VEO, Seedance, Vidu, Grok, and more — with durations from 4 to 30 seconds.
  </Card>

  <Card title="API Reference" icon="book" href="/api-reference/introduction">
    Browse the full API reference for every endpoint, request parameter, and response schema across all three API categories.
  </Card>
</CardGroup>

## Supported Models

### Video Models

<CardGroup cols={2}>
  <Card title="Sora" icon="play">
    OpenAI's video generation model. Variants include `openAiSora2`, `openAiSora2Stable`, `openAiSora2Plus`, and `sora2Stable`. Supports 4–15 second clips.
  </Card>

  <Card title="VEO" icon="play">
    Google's high-quality video generation model. Supports text-to-video and image-to-video with multiple durations and resolutions.
  </Card>

  <Card title="Seedance 1.5" icon="play">
    Volcano Engine's video model. Models: `seedance_1_5_pro_720p` and `seedance_1_5_pro_1080p`. Supports 5–10 second videos.
  </Card>

  <Card title="Seedance 2.0" icon="play">
    Next-generation Volcano Engine model. Variants: `seedance_2_0`, `seedance_2_0_fast`, `seedance_2_0_pro`, `seedance_2_0_fast_pro`. Includes asset management.
  </Card>

  <Card title="Vidu" icon="play">
    Flexible text-to-video and image-to-video model. Models: `viduq3-pro` and `viduq3-turbo`. Supports 1–16 seconds at 540p, 720p, or 1080p.
  </Card>

  <Card title="Grok" icon="play">
    xAI's video model. `grok_video3` supports 6–30 seconds at 480p/720p; `grok_video3_pro` produces fixed 10-second clips.
  </Card>

  <Card title="StarVideos" icon="play">
    High-quality video generation with text and image prompt support.
  </Card>

  <Card title="Videos & SD2 Manxue" icon="play">
    Additional specialized video generation models for diverse creative use cases.
  </Card>
</CardGroup>

### Image Models

| Model                                               | Description                                                                              |
| --------------------------------------------------- | ---------------------------------------------------------------------------------------- |
| `nano-banana` / `nano-banana-2` / `nano-banana-pro` | Nano Banana series — fast to professional quality, up to 4K resolution, 10 aspect ratios |
| `gpt-image-2`                                       | GPT Image 2 — high-fidelity image generation via the GPT Image endpoint                  |

### Text Models

| Model Family | Access Method                      | Example Models                                                       |
| ------------ | ---------------------------------- | -------------------------------------------------------------------- |
| GPT          | OpenAI-compatible Chat Completions | `gpt-5.4`, `gpt-5.5`                                                 |
| Claude       | OpenAI-compatible Chat Completions | `claude-sonnet-4-5`, `claude-opus-4-7`, `claude-haiku-4-5`           |
| Gemini       | OpenAI-compatible Chat Completions | `gemini-2.5-pro`, `gemini-3-flash-preview`, `gemini-3.1-pro-preview` |

## Base URLs

All GlobalAI OPC API requests go to one of two base URLs depending on the API category.

| API Category                                     | Base URL                                        |
| ------------------------------------------------ | ----------------------------------------------- |
| **Text** (GPT, Claude, Gemini)                   | `http://apillm.globalaiopc.com/gw_llm_power`    |
| **Image & Video** (Nano Banana, Sora, VEO, etc.) | `https://zcbservice.aizfw.cn/kyyReactApiServer` |

<Note>
  Always use the correct base URL for the API category you are calling. Using the text base URL for an image or video endpoint — or vice versa — will result in a connection error.
</Note>

## Authentication

All requests to GlobalAI OPC require a Bearer token: include your API key in an `Authorization: Bearer YOUR_API_KEY` header on every request.

## Async Task Model

Image and video generation use an **asynchronous task model**. When you submit a generation request, the API immediately returns a task `id` and a `status` of `queued` or `processing`. You then poll a query endpoint with the task ID until the status changes to `completed` (or `failed`), at which point the response includes the generated image URL or video URL.

<Note>
  Task completion time varies by model and output length. Video generation typically takes one to several minutes. Plan your integration to poll at reasonable intervals — every 30–60 seconds is a good starting point.
</Note>

## Next Steps

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/guides/quickstart">
    Make your first text, image, and video generation calls in minutes.
  </Card>

  <Card title="API Reference" icon="code" href="/api-reference/introduction">
    Explore the complete API reference for all endpoints and parameters.
  </Card>
</CardGroup>