API setup

The API is OpenAI-compatible: any client or library that lets you set a custom Base URL works.

Base URL

text

https://api.eni-cloud.xyz/v1

Authentication

Send the key in the Authorization header:

text

Authorization: Bearer eni-sk-...

Keys start with eni-sk-. We store only a hash of the key, so we can't show it again. Lost a key? Reissue it in the console; the old one stops working immediately.

Models

Put the ID of your key's model in the model field.

Model	`model` ID	Input, 1M tokens	Output, 1M tokens
DeepSeek	`deepseek`	$0.30	$1.20
Kimi	`kimi`	$0.30	$1.20
GLM	`glm`	$0.30	$1.20
Gemini	`gemini`	$0.30	$1.20

A key for one model doesn't work with another: the request returns 403 model_not_allowed.

Endpoints

Method	Path	What it does
POST	`/chat/completions`	Model reply, regular or streamed
GET	`/models`	The model available to this key

Streaming

Add "stream": true and the reply arrives in chunks (Server-Sent Events), just like OpenAI:

bash

curl https://api.eni-cloud.xyz/v1/chat/completions \
  -H "Authorization: Bearer eni-sk-..." \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek", "stream": true, "messages": [{"role": "user", "content": "Explain TCP"}]}'

Your own system prompt

You can send your own system message. The model's built-in prompt still goes first and yours follows. See How the built-in prompt works.