Model:

🔐

Account Required to Generate API Keys

To generate custom Millisa API Keys (mls_live_...), monitor your quotas, and connect apps, please sign in or create a free developer account.

Developer Integration Hub

API Version: v1 OpenAI & Gemini Unified

Complete architectural guides, copy-paste SDK implementations, and live endpoint testing for Web, Android, and iOS applications.

🚀 Quickstart Overview

Millisa is an intelligent AI Gateway that acts as a secure bridge between your client applications and upstream AI models (Google Gemini and OpenAI). Instead of hardcoding master API keys in client apps, you issue scoped Millisa API Keys (mls_live_...) with customized quotas and rate limits.

Base API URL

https://millisa.codly.in/api/v1

Authentication Scheme

Bearer mls_live_YOUR_KEY

Authentication Headers

All API requests must include your Millisa API Key in the Authorization HTTP header:

Standard HTTP Headers

Authorization: Bearer mls_live_YOUR_MILLISA_API_KEY
Content-Type: application/json

📦 Pre-Built SDK Clients

Download drop-in ready client files for your mobile or web projects:

🌐 Web (JavaScript / TypeScript)

Standard fetch, async/await, and Server-Sent Events (SSE) streaming reader.

📥 Download MillisaClient.js

🤖 Android (Kotlin)

Modern Kotlin with OkHttp, Coroutines, StateFlow, and Jetpack Compose hooks.

📥 Download MillisaClient.kt

🍎 iOS (Swift 5.5+)

Swift Concurrency (async/await), URLSession, and SwiftUI chat bindings.

📥 Download MillisaClient.swift

🌐 Web Integration Guide

Connect web applications built with Vanilla JavaScript, TypeScript, React, Next.js, or Vue.

1. Simple Agent Prompt Query

Send a single prompt and receive Millisa's response with persona and formatting applied:

JavaScript Fetch API

async function askMillisa(prompt) {
  const response = await fetch('https://millisa.codly.in/api/v1/agent/query', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'Authorization': 'Bearer mls_live_YOUR_MILLISA_API_KEY'
    },
    body: JSON.stringify({
      prompt: prompt,
      model: 'millisa-auto' // Options: 'millisa-auto', 'millisa-gemini-flash', 'millisa-gpt-4o-mini'
    })
  });

  const data = await response.json();
  if (!response.ok) {
    throw new Error(data.error?.message || 'Failed to query Millisa');
  }

  console.log('Millisa replied:', data.response);
  return data.response;
}

2. Real-Time Token Streaming (Server-Sent Events)

Stream Millisa's responses word-by-word in real time:

JavaScript SSE Stream Reader

async function streamMillisa(prompt, onChunk) {
  const res = await fetch('https://millisa.codly.in/api/v1/chat/completions', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'Authorization': 'Bearer mls_live_YOUR_MILLISA_API_KEY'
    },
    body: JSON.stringify({
      model: 'millisa-auto',
      messages: [{ role: 'user', content: prompt }],
      stream: true
    })
  });

  const reader = res.body.getReader();
  const decoder = new TextDecoder('utf-8');
  let buffer = '';

  while (true) {
    const { done, value } = await reader.read();
    if (done) break;
    buffer += decoder.decode(value, { stream: true });
    const lines = buffer.split('\n');
    buffer = lines.pop();

    for (const line of lines) {
      const trimmed = line.trim();
      if (!trimmed.startsWith('data:')) continue;
      const dataStr = trimmed.replace(/^data:\s*/, '');
      if (dataStr === '[DONE]') return;
      try {
        const json = JSON.parse(dataStr);
        const delta = json.choices?.[0]?.delta?.content;
        if (delta) onChunk(delta);
      } catch (e) {}
    }
  }
}

3. Drop-in OpenAI SDK Compatibility

You can use the official openai npm package directly by pointing the baseURL to Millisa:

Node.js / React with OpenAI SDK

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://millisa.codly.in/api/v1',
  apiKey: 'mls_live_YOUR_MILLISA_API_KEY',
  dangerouslyAllowBrowser: true // if testing in browser
});

const response = await client.chat.completions.create({
  model: 'millisa-auto',
  messages: [{ role: 'user', content: 'Explain quantum computing in simple terms.' }]
});

console.log(response.choices[0].message.content);

🤖 Android App Integration (Kotlin)

Comprehensive guide to integrate Millisa AI into Android apps using OkHttp, Kotlin Coroutines, and Jetpack Compose.

Step 1: Add Dependencies

In your app's build.gradle.kts:

build.gradle.kts

dependencies {
    implementation("com.squareup.okhttp3:okhttp:4.12.0")
    implementation("org.jetbrains.kotlinx:kotlinx-coroutines-android:1.8.0")
    implementation("com.google.code.gson:gson:2.10.1")
}

Step 2: Internet Permission

Ensure internet access is declared in AndroidManifest.xml:

AndroidManifest.xml

<uses-permission android:name="android.permission.INTERNET" />

Step 3: Android Client Implementation

MillisaClient.kt

package com.example.millisa

import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.withContext
import okhttp3.MediaType.Companion.toMediaType
import okhttp3.OkHttpClient
import okhttp3.Request
import okhttp3.RequestBody.Companion.toRequestBody
import org.json.JSONObject
import java.io.IOException

class MillisaClient(
    private val apiKey: String,
    private val baseUrl: String = "https://millisa.codly.in/api/v1"
) {
    private val client = OkHttpClient()
    private val jsonType = "application/json; charset=utf-8".toMediaType()

    /**
     * Send query to Millisa Agent
     */
    suspend fun query(prompt: String, model: String = "millisa-auto"): Result<String> {
        return withContext(Dispatchers.IO) {
            try {
                val payload = JSONObject().apply {
                    put("prompt", prompt)
                    put("model", model)
                }

                val request = Request.Builder()
                    .url("$baseUrl/agent/query")
                    .addHeader("Authorization", "Bearer $apiKey")
                    .addHeader("Content-Type", "application/json")
                    .post(payload.toString().toRequestBody(jsonType))
                    .build()

                val response = client.newCall(request).execute()
                val body = response.body?.string() ?: ""

                if (!response.isSuccessful) {
                    return@withContext Result.failure(IOException("HTTP ${response.code}: $body"))
                }

                val json = JSONObject(body)
                val reply = json.optString("response", "No response")
                Result.success(reply)
            } catch (e: Exception) {
                Result.failure(e)
            }
        }
    }
}

// ─────────────────────────────────────────────────────────────
// Usage in ViewModel / Activity:
// ─────────────────────────────────────────────────────────────
// val millisa = MillisaClient(apiKey = "mls_live_YOUR_KEY")
// lifecycleScope.launch {
//     val result = millisa.query("Hello Millisa, summarize today's tech news.")
//     result.onSuccess { text -> println(text) }
// }

🍎 iOS App Integration (Swift)

Modern iOS architecture using Swift Concurrency (async/await), URLSession, and SwiftUI. Compatible with iOS 15.0+ and macOS 12.0+.

Millisa iOS Client Implementation

MillisaClient.swift

import Foundation

public final class MillisaClient {
    private let apiKey: String
    private let baseURL: URL

    public init(apiKey: String, baseURL: String = "https://millisa.codly.in/api/v1") {
        self.apiKey = apiKey
        self.baseURL = URL(string: baseURL)!
    }

    /**
     * Send prompt query to Millisa
     */
    public func query(prompt: String, model: String = "millisa-auto") async throws -> String {
        let endpoint = baseURL.appendingPathComponent("agent/query")
        var request = URLRequest(url: endpoint)
        request.httpMethod = "POST"
        request.addValue("Bearer \(apiKey)", forHTTPHeaderField: "Authorization")
        request.addValue("application/json", forHTTPHeaderField: "Content-Type")

        let payload: [String: Any] = ["prompt": prompt, "model": model]
        request.httpBody = try JSONSerialization.data(withJSONObject: payload)

        let (data, response) = try await URLSession.shared.data(for: request)

        guard let httpResponse = response as? HTTPURLResponse, httpResponse.statusCode == 200 else {
            let errorText = String(data: data, encoding: .utf8) ?? "Server Error"
            throw NSError(domain: "Millisa", code: -1, userInfo: [NSLocalizedDescriptionKey: errorText])
        }

        guard let json = try JSONSerialization.jsonObject(with: data) as? [String: Any],
              let reply = json["response"] as? String else {
            throw URLError(.cannotParseResponse)
        }

        return reply
    }
}

// ─────────────────────────────────────────────────────────────
// Usage in SwiftUI View / ViewModel:
// ─────────────────────────────────────────────────────────────
// let client = MillisaClient(apiKey: "mls_live_YOUR_KEY")
// Task {
//     do {
//         let answer = try await client.query(prompt: "Hello Millisa from iOS!")
//         print("Answer: \(answer)")
//     } catch {
//         print("Error: \(error.localizedDescription)")
//     }
// }

POST /api/v1/chat/completions OpenAI Standard

Primary multi-turn chat completions endpoint. Fully compatible with the OpenAI API format and supports streaming Server-Sent Events.

Request Body Parameters:

Parameter	Type	Required	Description
messages	array	Required	Array of message objects with `role` ("user", "assistant", "system") and `content` string.
model	string	Optional	Model to use: `millisa-auto` (default), `millisa-gemini-flash`, `millisa-gemini-pro`, `millisa-gpt-4o-mini`, `millisa-gpt-4o`.
stream	boolean	Optional	Set to `true` to stream response tokens chunk-by-chunk via SSE.
temperature	float	Optional	Sampling temperature between 0.0 (focused) and 1.0 (creative). Default: `0.7`.
max_tokens	integer	Optional	Maximum tokens to generate. Default: `2048`.

Sample Request (cURL):

cURL Request

curl -X POST https://millisa.codly.in/api/v1/chat/completions \
  -H "Authorization: Bearer mls_live_YOUR_MILLISA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "millisa-auto",
    "messages": [
      {"role": "system", "content": "You are a helpful coding assistant."},
      {"role": "user", "content": "Write a binary search function in Python."}
    ],
    "temperature": 0.7
  }'

POST /api/v1/agent/query Simplified Format

Lightweight single-prompt endpoint designed for quick queries and mobile apps.

Request Body Parameters:

Parameter	Type	Required	Description
prompt	string	Required	The text question or instruction for Millisa AI.
model	string	Optional	Target model ID. Defaults to `millisa-auto`.

Sample Response (JSON):

Response 200 OK

{
  "success": true,
  "agent": "Millisa",
  "model": "gemini-1.5-flash",
  "provider": "gemini",
  "response": "Hello! I am Millisa, your dedicated AI assistant. How can I help you today?",
  "usage": {
    "prompt_tokens": 18,
    "completion_tokens": 22,
    "total_tokens": 40
  },
  "latency_ms": 320
}

GET /api/v1/models

Returns a list of all models available under your Millisa Gateway.

GET /api/v1/usage

Returns your active Millisa API key metrics, remaining monthly quota, and rate limits.

🧪 Interactive API Playground

Test your live Millisa API Key directly against the gateway right here in the browser.

Select Key or Enter Raw Key:

API Endpoint:

AI Model:

Prompt:

Live Response Payload

Status: Ready 0 ms

Click "Send Live Request" to test your API key.

🛡️ HTTP Status Codes & Error Handling

Millisa Gateway uses standard HTTP response codes to indicate request success or failure.

Status Code	Error Type	Meaning & Troubleshooting
200 OK	Success	Request succeeded and the response payload is returned.
400 Bad Request	invalid_request_error	Missing required parameters (e.g. missing `prompt` or `messages` array).
401 Unauthorized	authentication_error	Missing or invalid Millisa API key. Verify header `Authorization: Bearer mls_live_...`.
403 Forbidden	permission_error	The Millisa API key has been revoked or suspended by the developer/admin.
429 Rate Limit	rate_limit_exceeded	Requests per minute limit exceeded or monthly quota reached. Implement backoff retry.
500 Server Error	api_error	Upstream provider (OpenAI or Gemini) error or configuration issue.

Rate Limiting Policy

Millisa applies a sliding 60-second window per API key. If the request rate exceeds your configured RPM (default: 60 requests/min), the gateway will return HTTP 429. Exponential backoff retry logic is recommended for mobile client integrations.

Developer User Management

Super Admin View: View all registered developers, monitor API keys issued, and manage account statuses.

Registered Developers

Developer	Role	Keys Issued	Total Calls	Status	Actions

Super Admin Master Key Vault

Securely configure your master OpenAI and Google Gemini API keys. These keys are kept safely on the server and are never revealed to developers or client apps.

🔑 Master AI Provider Keys

Google Gemini API Key: Obtain from Google AI Studio. Used for Gemini 1.5 Flash & Pro models.

OpenAI API Key: Obtain from OpenAI Platform. Used for GPT-4o & GPT-4o-mini models.

Default Model Provider:

🤖 Millisa Agent Persona & System Prompt

Agent Display Name:

System Persona Prompt: This prompt defines Millisa's behavior, personality, and tone across all endpoints.

⚙️ Defaults & Security

Platform Name:

Default Rate Limit (RPM):

Default Monthly Quota:

Admin Password (Leave blank to keep):

Account Required to Generate API Keys

My Millisa API Keys

Developer Integration Hub

🚀 Quickstart Overview

Authentication Headers

📦 Pre-Built SDK Clients

🌐 Web (JavaScript / TypeScript)

🤖 Android (Kotlin)

🍎 iOS (Swift 5.5+)

🌐 Web Integration Guide

1. Simple Agent Prompt Query

2. Real-Time Token Streaming (Server-Sent Events)

3. Drop-in OpenAI SDK Compatibility

🤖 Android App Integration (Kotlin)

Step 1: Add Dependencies

Step 2: Internet Permission

Step 3: Android Client Implementation

🍎 iOS App Integration (Swift)

Millisa iOS Client Implementation

Request Body Parameters:

Sample Request (cURL):

Request Body Parameters:

Sample Response (JSON):

🧪 Interactive API Playground

🛡️ HTTP Status Codes & Error Handling

Rate Limiting Policy

Developer User Management

Registered Developers

Super Admin Master Key Vault

🔑 Master AI Provider Keys

🤖 Millisa Agent Persona & System Prompt

⚙️ Defaults & Security

Developer Portal

Create Millisa API Key

Key Generated Successfully!