​
Authentication

​
Headers

Two headers are required for authentication:

  1. apikey

    • Description: Your unique API key to access the endpoint. This key is specific to your organization or project and can be found in your Humiris dashboard Here.

  2. mixTuningId

    • Description: The unique identifier for your MixTuning configuration (Mix Model).

​
Endpoint: Chat Completions

​
URL

Base URL : https://moai-service-app.humiris.ai/api/v1

POST /chat/completions

​
Description

Use this endpoint to send a chat completion request. This is similar to OpenAI’s chat/completions endpoint, with added functionality for custom tuning.

​
Headers

HeaderTypeRequiredDescription
apikeystringYesYour API key.
mixTuningIdstringYesThe MixTuning ID for personalized AI responses.

​
Body

The request body follows OpenAI’s chat/completions format.

​
Request Body Example

{
  "messages": [
    {
      "role": "user",
      "content": "Say this is a test."
    }
  ],
  "model": "Humiris/humiris-moai"
}
FielsTypeRequiredDescription
messagesstringYesArray of message objects, each with role and content.
modelstringYesThe model to use. Example: Humiris/humiris-moai

Response

{
  "id": "chatcmpl-12345",
  "object": "chat.completion",
  "created": 1698000000,
  "model": "Humiris/humiris-moai",
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "This is a test."
      },
      "finish_reason": "stop",
      "stop_reason": "\n"
    }
  ],
  "usage": {
    "prompt_tokens": 5,
    "completion_tokens": 7,
    "total_tokens": 12
  }
}

Error

Status CodeErrorDescription
400BadRequestThe mixTuningId header is missing.
404MixTuningNotFoundThe provided mixTuningId does not exist.
401UnauthorizedInvalid or missing apikey.

​
Integration Examples

JavaScript Here’s how to integrate the API using the fetch method in JavaScript:

const fetch = require("node-fetch");

const apikey = "your-api-key";
const mixTuningId = "your-mix-tuning-id";

async function sendChatCompletion() {
  const response = await fetch(
    "https://moai-service-app.humiris.ai/api/v1/chat/completions",
    {
      method: "POST",
      headers: {
        "Content-Type": "application/json",
        apikey: apikey,
        mixTuningId: mixTuningId,
      },
      body: JSON.stringify({
        messages: [
          {
            role: "user",
            content: "Say this is a test.",
          },
        ],
        model: "Humiris/humiris-moai",
      }),
    }
  );

  if (!response.ok) {
    const error = await response.json();
    console.error("Error:", error);
    return;
  }

  const data = await response.json();
  console.log("Response:", data);
}

sendChatCompletion();

Python Here’s how to integrate the API using the request method in Python:

import requests

url = "https://moai-service-app.humiris.ai/api/v1/chat/completions"
headers = {
    "Content-Type": "application/json",
    "apikey": "your-api-key",
    "mixTuningId": "your-mix-tuning-id"
}
payload = {
    "messages": [
        {
            "role": "user",
            "content": "Say this is a test."
        }
    ],
    "model": "Humiris/humiris-moai"
}

response = requests.post(url, json=payload, headers=headers)

if response.status_code == 200:
    print("Response:", response.json())
else:
    print("Error:", response.status_code, response.json())

Support For any questions or issues, please contact Hilario Houmey at h.hilario@humiris.ai.

Python