API Documentation

OpenAI-compatible. Call any model in 3 steps.

Call any model with one Base URL

Point your OpenAI-compatible client or SDK directly at this gateway. Use the unified Base URL below with an API key generated from this dashboard to securely proxy LLM requests.

OpenAI compatibleRESTStreaming

URL cơ sở

https://your-domain.com/v1

Authorization header

Authorization: Bearer YOUR_API_KEY

System health

Live status of the gateway and upstream services.

Gateway

Checking…

Pinging base URL…

Down

Version

—

your-domain.com/v1

Thời gian hoạt động

—

Không rõ

Status endpoint

Reachable

GET /api/status

Quick Start Guide

Get from zero to your first request in three steps.

1. Get an API key

Create a token in the API Keys tab. Treat this key like a password — it allows authorization for all downstream models.

Go to API Keys Top up balance

Security Warning

Never commit API keys to GitHub or expose them in client-side code. Always load keys via environment variables on a secure backend.

2. Pick a model

Choose from the enabled models in your group. Clicking a model tag updates the snippets below automatically.

Browse all models →

Chat Completions

POST/v1/chat/completions

Replace $YOUR_API_KEY with the token you generated, select a language tab, and execute the call.

chat-completions.sh

bash

curl https://your-domain.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ]
  }'

Streaming responses

Set "stream": true in the payload request to stream token completions instantly using Server-Sent Events (SSE).

stream-completions.sh

bash

curl https://your-domain.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -d '{
    "model": "gpt-4o-mini",
    "stream": true,
    "messages": [
      {"role": "user", "content": "Stream a short greeting."}
    ]
  }'

Embeddings

POST/v1/embeddings

Calculate embedding vectors for text inputs using standard compatible formats.

embeddings.sh

bash

curl https://your-domain.com/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -d '{
    "model": "text-embedding-3-small",
    "input": "The quick brown fox"
  }'

Tạo hình ảnh

POST/v1/images/generations

Generate images from text descriptions using the DALL-E compatible endpoint.

image-generation.sh

bash

curl https://your-domain.com/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "a cat reading a book, watercolor",
    "size": "1024x1024"
  }'

Submit standard JSON parameters including size, response format, and model fields to customize generation output.

Common Errors

Errors match the standard OpenAI error format, allowing integrations with existing SDKs to handle exceptions smoothly.

Trạng thái	Meaning	How to fix
401	Invalid API key	Check the Authorization header format and confirm key is active.
402	Insufficient balance	Top up your wallet balance or check active credit limits.
404	Không tìm thấy mô hình	Ensure model name is correct, active, and accessible to your user group.
429	Rate limited	Slow down requests rate, or contact the admin to request high limits.
5xx	Upstream error	The provider/upstream service is down. Retry with exponential backoff.