Welcome to WS AI Gateway Developer Documentation¶

WS AI Gateway

Enterprise-grade Unified LLM Interface Platform

About WS AI Gateway¶

WS AI Gateway is an enterprise-grade unified interface platform for Large Language Models (LLMs). It aggregates 50+ mainstream AI models — Claude, GPT, Gemini, DeepSeek, GLM, Kimi, Qwen, Grok and more — behind a single endpoint, with unified authentication, billing, and monitoring.

Base URL¶

All API requests use the following base URL:

http://baseurl

Note

http://baseurl is a placeholder. Please contact your company's sales manager to obtain the actual base URL.

Authentication¶

Every request must carry your API key. For OpenAI-compatible endpoints, use the Authorization header:

Authorization: Bearer $WS_API_KEY

Where $WS_API_KEY is the API key created in the console (starts with sk-).

Quick Start¶

Send your first chat request:

curl http://baseurl/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WS_API_KEY" \
  -d '{
    "model": "gpt-5.4-mini",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

The gateway also natively supports the Anthropic Messages format (/v1/messages) and the Google Gemini format (/v1beta/models/{model}:generateContent) — use whichever SDK or format your application already speaks.

Model Catalog

Browse all available models, grouped by capability, with the API formats each model supports.

View Models
API Reference

Detailed endpoint definitions, authentication, request parameters, and response examples.

View API
Image & Video Generation

Generate images and videos with GPT-Image, Gemini, Veo, Kling and Vidu models.

View Image API
Usage Query

Query the quota usage of your API key at any time.

View Usage API