Documentation v2.4

Intelligence
Infrastructure.

Welcome to the Ravonix API documentation. Build the next generation of AI-powered Minecraft applications with our unified infrastructure.

Base URL

https://api.ravonixapi.com/v1

Authentication

All Ravonix API requests must be authenticated using an API key via the Authorization HTTP header using the Bearer token format.

Authorization: Bearer YOUR_API_KEY

Your API key identifies your account and is used to track usage, enforce rate limits, and secure your requests. Never share your API key publicly.

Core Control Settings

To build a truly immersive Minecraft experience, you need to control how the AI thinks and speaks. These six settings act as the "knobs and dials" of your intelligence engine. Because Ravonix is a unified platform, these parameters work identically whether you are using our Spark, Forge, or Void models.

1. Temperature (Randomness)

"temperature": 0.7

Temperature controls the "creativity" or "predictability" of the AI. Think of it as a scale between a Strict Robot and a Wild Artist.

Low (0.0 to 0.3): The AI will always choose the most likely next word. This is perfect for technical help, Spigot API queries, or NMS logic where you need the exact correct answer every time.
Medium (0.7): This is our default setting. It provides a balance between being smart and being interesting. Great for general chat and basic NPCs.
High (1.2 to 2.0): The AI takes risks. It will use rare words and complex sentence structures. Use this for drunk pirates in a tavern, mysterious ancient gods, or creative world lore.

2. Personality (Global Persona)

"personality": "A grumpy old miner"

The personality field allows you to set a global identity for the AI that persists across the conversation. This is perfect for defining how an NPC should act, their emotional state, or their specific speech patterns.

Instead of putting the identity in the main prompt every time, you can set it once in this field. For example, if you set it to "A talkative pirate," the AI will automatically use words like "Ahoy," "Matey," and "Scallywag" without you needing to tell it to in your message.Pro Tip: Use this in combination with Temperature to make your game characters truly unique.

3. Max Tokens (Length)

"max_tokens": 150

This setting tells the AI exactly when to stop talking. In Minecraft, the chat box has limited space, and player attention spans are short. Managing tokens saves you money and keeps the game fast.

One "token" is roughly 4 characters or 0.75 words. If you set this to 50, the AI will give you about 1-2 sentences. If you set it to 500, it can write an entire chapter for a Minecraft book.Pro Tip: Always set a limit to prevent the AI from rambling and wasting your token balance.

4. Top P (Vocabulary Diversity)

"top_p": 1.0

Top P is a more advanced version of Temperature. It tells the AI to only consider the "top" percentage of words that make sense in that moment.

If you set Top P to 0.1, the AI will only look at the top 10% of logical words. This makes the AI very focused and consistent. If you leave it at 1.0, it considers every possible word in its vocabulary. We recommend leaving this at 1.0 and using Temperature to adjust creativity instead.

5. Stop Sequences (The Emergency Brake)

"stop": ["\n", "END"]

A Stop Sequence is a specific word or character that tells the AI to immediately shut up.

For example, if you are building a chat bot for a Minecraft server and you don't want the AI to start a new paragraph, you can set the stop sequence to \n (a new line). As soon as the AI tries to hit "Enter," it will stop and send whatever it has already typed. This is essential for keeping NPC dialogue clean and formatted correctly for the game chat.

6. N (Brainstorming Mode)

"n": 1

The "N" parameter tells the AI to generate multiple different answers to the same question at the exact same time.

If you are a Game Designer and you need 3 different ideas for a "Quest Reward," you can set n=3. The API will return a list of 3 unique ideas. You can then let your code pick the best one or show all three to the player. Note: This uses more tokens because you are receiving multiple responses.

Developer Quickstart

Java (Paper/Spigot)

Maven Dependency

<dependency>
    <groupId>com.ravonixapi</groupId>
    <artifactId>ravonix-sdk</artifactId>
    <version>1.0.0</version>
</dependency>

Implementation

import ravonixapi.Ravonix;
import ravonixapi.models.RavonixResponse;

Ravonix ravonix = new Ravonix("Rit_YOUR_KEY");

ravonix.chat().ask("Spark", "Hello!").thenAccept(res -> {
    System.out.println(res.getOutput());
});

JavaScript (Bedrock)

NPM Install

npm install @ravonixapi/sdk

Implementation

import { Ravonix } from "@ravonixapi/sdk";

const ravonix = new Ravonix("Rit_YOUR_KEY");

ravonix.chat().ask("Spark", "Hello!").then(res -> {
    console.log(res.output);
});