Quick Start with JuiceCore

This section will help you get started with JuiceCore as quickly as possible without unnecessary details. It's written to be understandable even for those working with AI APIs for the first time, while remaining accurate and useful for professional developers.

In short: get a key → make a request → get a response.

🧩 What is JuiceCore (briefly)

JuiceCore is an OpenAI-compatible API that automatically selects the optimal AI model for your task. You don't need to choose which specific model works "under the hood" — just select the logical type:

• JuiceAi-Fast — fast and cheap
• JuiceAi-Pro — maximum quality and intelligence
• JuiceAi-Coder — specifically for code

📋 What You'll Need

Before starting, make sure you have:

1. JuiceCore API key — issued after registration
2. API Base URL — always the same:

https://api.juicecore.xyz/v1
3. Any HTTP client — cURL, Postman, fetch, axios, OpenAI SDK, etc.

🔑 Step 1: Getting an API Key

1. Go to juicecore.xyz
2. Register or log into your account
3. Open the API Keys section
4. Click Create new key
5. Copy the generated key

Example key:

jc_xxxxxxxxxxxxxxxxxxxxxxxx

⚠️ Security:

• never store the API key in frontend code
• don't commit it to GitHub
• use environment variables (ENV)

🚀 Step 2: First API Request

The simplest way to verify everything works is to make a request via cURL.

Example: simple chat

curl https://api.juicecore.xyz/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "JuiceAi-Fast",
    "messages": [
      {
        "role": "user",
        "content": "Hello! How are you?"
      }
    ]
  }'

🔁 Be sure to replace YOUR_API_KEY with your actual key.

📦 What JuiceCore Returns

The response comes in standard OpenAI API format — as JSON.

Example shortened response:

{
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Hello! I'm doing great. How can I help you?"
      }
    }
  ],
  "usage": {
    "total_tokens": 32
  }
}

What's Important Here

• message.content — the main AI response text
• usage.total_tokens — how many tokens were used
• the format is always stable, regardless of which model was selected inside JuiceCore

🧠 Step 3: Choosing the Right Model

You don't choose a specific AI model, only the task type.

💻 Example: Code Generation

curl https://api.juicecore.xyz/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "JuiceAi-Coder",
    "messages": [
      {
        "role": "user",
        "content": "Write a Python function to sort a list"
      }
    ]
  }'

JuiceCore will automatically select the optimal model for working with code and return the result in standard format.

✅ Done

At this point, you can already:

• integrate JuiceCore into your application
• replace OpenAI API without rewriting logic
• use AI without being tied to a single provider

📘 What's Next

We recommend moving to the following sections:

• Authentication — more details about keys and limits
• Models & Routing — how internal model selection works
• SDK Examples — examples for JavaScript, Python, and other languages

JuiceCore is designed to be simple at the start and powerful in production.