MultiConnection API Documentation
MultiConnection provides an OpenAI-compatible API. Keep your existing SDK and replace the base URL and API key to route requests through MultiConnection with billing and usage records.
Base URL and Authentication
Base URL: https://multiconnection.cloud/v1
All /v1/* requests use a Bearer token:
Authorization: Bearer sk-live-...
curl Example
curl https://multiconnection.cloud/v1/chat/completions \
-H "Authorization: Bearer sk-live-..." \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-flash",
"messages": [
{ "role": "user", "content": "hello MultiConnection" }
]
}'
OpenAI JavaScript SDK
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.MULTICONNECTION_API_KEY,
baseURL: "https://multiconnection.cloud/v1"
});
const completion = await client.chat.completions.create({
model: "deepseek-v4-flash",
messages: [{ role: "user", content: "hello MultiConnection" }]
});
console.log(completion.choices[0].message.content);
OpenAI Python SDK
from openai import OpenAI
import os
client = OpenAI(
api_key=os.environ["MULTICONNECTION_API_KEY"],
base_url="https://multiconnection.cloud/v1",
)
completion = client.chat.completions.create(
model="deepseek-v4-flash",
messages=[{"role": "user", "content": "hello MultiConnection"}],
)
print(completion.choices[0].message.content)
Model List
The console and homepage pricing table read from /api/models. The OpenAI-compatible model list is GET /v1/models and requires a user API key.
Streaming
stream=true is supported and returns Server-Sent Events.
curl https://multiconnection.cloud/v1/chat/completions \
-H "Authorization: Bearer sk-live-..." \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-flash",
"stream": true,
"messages": [
{ "role": "user", "content": "stream test" }
]
}'
Error Codes
| code | Meaning |
|---|---|
api_key_required | Missing user API key. |
invalid_api_key | The API key is invalid, deleted or disabled. |
insufficient_balance | The account balance is too low. Recharge or ask for test credit. |
model_not_found | The model does not exist or is not enabled. |
rate_limited | The API key or IP rate limit was exceeded. |
upstream_error | An upstream provider returned an error. |
admin_forbidden | Admin permission is required. |
Recharge and Test Credit
New accounts may receive a small test balance to verify API keys, model calls, billing and usage records. When balance is low, submit a recharge order in the console and wait for admin approval.
Enterprise transfer and contract customers can work with the Singapore entity: MULTIVERSE CONNECTION TECHNOLOGY PTE. LTD.. Full bank account details are provided in the console recharge page or confirmed by an administrator.