Skip to main content

Documentation Index

Fetch the complete documentation index at: https://xingchao.mintlify.app/llms.txt

Use this file to discover all available pages before exploring further.

The GlobalAI OPC API provides REST endpoints for generating text, images, and videos using state-of-the-art AI models. All endpoints use Bearer token authentication. Whether you are building conversational applications, generating creative imagery, or producing AI-driven video content, this reference covers every endpoint, parameter, and response format you need.

​
Base URLs

Use the appropriate base URL for the API category you are calling: 
Text API:           http://apillm.globalaiopc.com/gw_llm_power
Image & Video API:  https://zcbservice.aizfw.cn/kyyReactApiServer

​
Authentication

Include your API key as a Bearer token on every request: 
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
Replace YOUR_API_KEY with the key issued to your account. Contact GlobalAI OPC support if you need to obtain or rotate a key.

​
API Categories

Text APIs

OpenAI-compatible chat, Claude Messages, and Gemini native endpoints for intelligent conversation and text generation.

Image APIs

Nano Banana and GPT Image 2 endpoints for high-quality image generation from text prompts.

Video APIs

Nine video model families — including Sora, VEO, Seedance, Vidu, and Grok — for text-to-video and image-to-video generation.

Asset Management

Upload and manage reference media assets used as inputs for image-to-video and guided generation workflows.

​
Video Generation Models

The table below summarises the available video models, their creation endpoints, maximum output resolution, and supported duration range.
ModelEndpointMax ResolutionDuration
SoraPOST /v1/sora/videos1280×7204–12s
VEOPOST /v1/veo/videos4Kfixed
Seedance 1.5POST /v1/seedance/videos1080p2–12s
Seedance 2.0POST /v1/kyyvideo2/videosvaries4–15s
ViduPOST /v1/vidu/videos1080p1–16s
GrokPOST /v1/grok/videos720p6–30s
StarVideosPOST /v1/starvideos/videosvaries3–15s
VideosPOST /v1/videos/videos720p4–15s
SD2 ManxuePOST /v1/sd2_manxue/videos4K4–15s

​
Async Task Flow

Video and image generation are asynchronous. Follow these three steps after making a creation request:
  1. Create the task — send a POST request to the relevant endpoint. The response returns a task id.
  2. Poll for status — send GET /v1/result/{id} repeatedly and inspect the status field in the response.
  3. Retrieve the result — when status is completed, read the video_url or image_url field to access your generated asset.

​
Common Response Statuses

Every task query response includes a status field with one of the following values:
StatusDescription
queuedWaiting to be processed
processingCurrently generating
completedDone — result URL available
failedGeneration failed — check the error field for details

​
Next Steps

Quickstart

Get your first API call working in minutes with step-by-step instructions and copy-ready code samples.

Authentication

Learn how to obtain your API key, set up Bearer token headers, and manage credentials securely.