DOCUMENTATION · OpenAI-compatible

API
Dokumentasi

MecutinAI adalah gateway AI OpenAI-compatible. Beli token pass, dapat API key, langsung pakai. Semua model dari provider AI terdepan.

Quick Start

Base URL

endpoint

https://mecutinai.com/api/v1

Autentikasi

Semua endpoint gateway memerlukan API key. Gunakan header Authorization dengan format Bearer. API key diawali dengan sk_.

auth

Authorization: Bearer sk_aB3xK9pQxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Request Pertama Anda

curl

curl -X POST https://mecutinai.com/api/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_AKUN_ANDA" \
  -d '{
    "model": "qwen3.7-max",
    "messages": [
      { "role": "user", "content": "Halo, apa kabar?" }
    ]
  }'

Model yang Tersedia

Semua model berasal dari provider AI terdepan. Katalog ini tetap — tidak berubah saat admin mengganti endpoint/key provider.

Model ID	Nama	Tipe	Deskripsi
qwen3.7-max	Qwen 3.7 Max	Chat	Flagship — reasoning terkuat, 1M context
qwen3.7-plus	Qwen 3.7 Plus	Chat	Balance performa & kecepatan, 1M context
qwen3.8-max	Qwen 3.8 Max	Chat	Generasi terbaru — flagship reasoning
deepseek-v4-pro	DeepSeek V4 Pro	Chat	Reasoning model — deep thinking
deepseek-v4-flash	DeepSeek V4 Flash	Chat	Cepat & hemat untuk tugas ringan
deepseek-v4-flash-0731	DeepSeek V4 Flash (0731)	Chat	Snapshot 31 Juli — versi tertentu
glm-5.1	GLM 5.1	Chat	Chat & function calling
glm-5.2	GLM 5.2	Chat	Generasi terbaru — chat & tools
kimi-k2.7-code	Kimi K2.7 Code	Chat	Spesialis coding & long context

Chat Completions

POST/v1/chat/completions

Menghasilkan respons chat dari model. Mendukung streaming (SSE) dan non-streaming. Format request & response 100% OpenAI-compatible.

Parameter Body

Field	Tipe	Wajib	Deskripsi
model	string	Ya	ID model (lihat tabel di atas)
messages	array	Ya	Array of { role, content }
stream	boolean	Tidak	Default false. true = SSE stream
temperature	number	Tidak	0.0–2.0, default 1.0
max_tokens	integer	Tidak	Maksimum token output
tools	array	Tidak	Function calling (model-dependent)

Contoh — Non-Streaming (curl)

curl

curl -X POST https://mecutinai.com/api/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_AKUN_ANDA" \
  -d '{
    "model": "qwen3.7-max",
    "messages": [
      { "role": "system", "content": "Kamu adalah asisten yang membantu." },
      { "role": "user", "content": "Jelaskan apa itu API gateway." }
    ],
    "temperature": 0.7
  }'

Response (Non-Streaming)

json

{
  "id": "chatcmpl-xxxxx",
  "object": "chat.completion",
  "model": "qwen3.7-max",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "API gateway adalah..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 25,
    "completion_tokens": 150,
    "total_tokens": 175
  }
}

Contoh — Streaming (SSE)

curl

curl -X POST https://mecutinai.com/api/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_AKUN_ANDA" \
  -d '{
    "model": "qwen3.7-plus",
    "messages": [{ "role": "user", "content": "Halo!" }],
    "stream": true
  }'

Response berupa Server-Sent Events. Setiap chunk berisi delta.content. Stream diakhiri dengan data: [DONE].

sse

data: {"choices":[{"delta":{"role":"assistant","content":""},"index":0}]}

data: {"choices":[{"delta":{"content":"Halo"},"index":0}]}

data: {"choices":[{"delta":{"content":"! Ada"},"index":0}]}

data: {"choices":[],"usage":{"prompt_tokens":5,"completion_tokens":10,"total_tokens":15}}

data: [DONE]

Contoh — JavaScript (fetch)

const res = await fetch('https://mecutinai.com/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer sk_AKUN_ANDA',
  },
  body: JSON.stringify({
    model: 'qwen3.7-max',
    messages: [{ role: 'user', content: 'Halo!' }],
  }),
});
const data = await res.json();
console.log(data.choices[0].message.content);

Contoh — Python (openai SDK)

python

from openai import OpenAI

client = OpenAI(
    api_key="sk_AKUN_ANDA",
    base_url="https://mecutinai.com/api/v1",
)

response = client.chat.completions.create(
    model="qwen3.7-max",
    messages=[{"role": "user", "content": "Halo!"}],
)

print(response.choices[0].message.content)

List Models

GET/v1/models

Mendapatkan daftar semua model yang tersedia. Format OpenAI-compatible.

curl

curl https://mecutinai.com/api/v1/models \
  -H "Authorization: Bearer sk_AKUN_ANDA"

json

{
  "object": "list",
  "data": [
    { "id": "qwen3.7-max", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    { "id": "qwen3.7-plus", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    { "id": "qwen3.8-max", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    { "id": "deepseek-v4-pro", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    { "id": "deepseek-v4-flash", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    { "id": "glm-5.2", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    { "id": "kimi-k2.7-code", "object": "model", "created": 1754870400, "owned_by": "mecutinai" },
    ...
  ]
}

Embeddings

POST/v1/embeddings

Endpoint embeddings tersedia jika provider mendukung model embedding. Cek /v1/models untuk model yang tersedia.

curl

curl -X POST https://mecutinai.com/api/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_AKUN_ANDA" \
  -d '{
    "model": "text-embedding-v3",
    "input": "Ini adalah teks yang akan di-embed"
  }'

json

{
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "index": 0,
      "embedding": [0.0123, -0.0456, 0.0789, ...]
    }
  ],
  "model": "text-embedding-v3",
  "usage": { "prompt_tokens": 8, "total_tokens": 8 }
}

Usage

GET/v1/usage

Cek sisa kuota token, status plan, dan statistik pemakaian.

curl

curl https://mecutinai.com/api/v1/usage \
  -H "Authorization: Bearer sk_AKUN_ANDA"

Response (Plan Terbatas)

json

{
  "plan": "HARIAN_10M",
  "display_name": "10M Tokens (1 Hari)",
  "unlimited": false,
  "token_limit": 10000000,
  "remaining_tokens": 8420000,
  "used_tokens": 1580000,
  "rpm_limit": 60,
  "expires_at": "2026-08-12T14:30:00Z",
  "time_remaining_hours": 18.5,
  "usage": { "requests": 42, "prompt_tokens": 510000, "completion_tokens": 1070000, "total_tokens": 1580000 }
}

Response (Plan Unlimited)

json

{
  "plan": "HARIAN_UNLIMITED",
  "display_name": "Unlimited Tokens (1 Hari)",
  "unlimited": true,
  "token_limit": null,
  "remaining_tokens": null,
  "used_tokens": 2500000,
  "rpm_limit": 15,
  "expires_at": "2026-08-12T14:30:00Z",
  "time_remaining_hours": 18.5,
  "usage": { "requests": 120, "prompt_tokens": 800000, "completion_tokens": 1700000, "total_tokens": 2500000 }
}

Rate Limiting & Kuota

Setiap API key memiliki batas RPM (requests per minute) dan kuota token berdasarkan plan yang dibeli. Rate limit di-enforced per API key menggunakan sliding window 60 detik.

Plan	Kuota Token	Durasi	RPM	Harga
10M Tokens (1 Hari)	10.000.000	24 jam	60	Rp5.000
Unlimited (1 Hari)	Tanpa batas	24 jam	15	Rp25.000
10M Tokens	10.000.000	Tanpa batas waktu	60	Rp20.000

Response Headers

Header	Deskripsi
X-RateLimit-Remaining-Requests	Sisa request dalam window 60 detik
X-RateLimit-Remaining-Tokens	Sisa token (di-omit untuk plan unlimited)
X-Plan-Expires-At	Waktu kedaluwarsa plan (di-omit untuk usage-based)
X-MecutinAI-Model	Model yang dipakai pada request ini

Kode Error

Semua error mengikuti format OpenAI.

json

{
  "error": {
    "message": "Deskripsi error",
    "type": "error_type",
    "code": "error_code"
  }
}

HTTP	Type	Code	Trigger
401	authentication_error	invalid_api_key	API key salah/revoked
403	permission_error	plan_expired	Plan kedaluwarsa
403	permission_error	insufficient_quota	Kuota token habis
429	rate_limit_error	rate_limit_exceeded	RPM terlampaui
400	invalid_request_error	invalid_model	Model tidak ada di katalog
400	invalid_request_error	invalid_request	Body request invalid
503	service_unavailable	provider_not_configured	Provider belum dikonfigurasi
502	api_error	provider_error	Upstream provider error
504	api_error	provider_timeout	Upstream timeout (120s)

SDK Compatibility

MecutinAI 100% OpenAI-compatible. Ganti base_url dan api_key:

python

# Python
from openai import OpenAI

client = OpenAI(
    api_key="sk_AKUN_ANDA",
    base_url="https://mecutinai.com/api/v1",
)

# LangChain
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
    openai_api_key="sk_AKUN_ANDA",
    openai_api_base="https://mecutinai.com/api/v1",
    model="qwen3.7-max",
)

javascript

// JavaScript / TypeScript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk_AKUN_ANDA",
  baseURL: "https://mecutinai.com/api/v1",
});

IDE & CLI Integration

MecutinAI adalah OpenAI-compatible, jadi bisa dipakai di berbagai IDE dan CLI coding assistant. Cukup ganti Base URL dan API Key di pengaturan masing-masing tool.

Base URL: https://mecutinai.com/api/v1
API Key: sk_AKUN_ANDA (dapatkan di /pricing)

🤖 Claude CLI (Claude Code)

Claude CLI mendukung custom OpenAI-compatible endpoint via environment variable.

bash

# Set environment variables
export ANTHROPIC_BASE_URL="https://mecutinai.com/api"
export ANTHROPIC_API_KEY="sk_AKUN_ANDA"

# Jalankan Claude CLI
claude

Catatan: Base URL tidak perlu trailing /v1 — Claude CLI menambahkan path secara internal.

🖥 Cursor IDE

Cursor mendukung override OpenAI Base URL untuk model OpenAI-compatible.

1. Buka Settings → Models
2. Expand API Keys section
3. Toggle ON Override OpenAI Base URL
4. Masukkan Base URL: https://mecutinai.com/api/v1
5. Masukkan API Key: sk_AKUN_ANDA

Mode Ask dan Plan bekerja dengan custom base URL. Agent mode memerlukan model yang support tool calling.

📝 Cline (VS Code Extension)

Cline support OpenAI-compatible provider dengan custom Base URL.

1. Buka VS Code → Ctrl+Shift+P → Cline: Open Settings
2. API Provider: pilih OpenAI Compatible
3. Base URL: https://mecutinai.com/api/v1
4. API Key: sk_AKUN_ANDA
5. Model: pilih model dari katalog (mis. deepseek-v4-flash)

⚡ Continue.dev (VS Code / JetBrains)

Continue.dev support custom OpenAI-compatible provider via config.json.

config.json (~/.continue/config.json)

{
  "models": [
    {
      "title": "MecutinAI DeepSeek",
      "provider": "openai",
      "model": "deepseek-v4-flash",
      "apiBase": "https://mecutinai.com/api/v1",
      "apiKey": "sk_AKUN_ANDA"
    }
  ]
}

🦀 Zed Editor

Zed support OpenAI-compatible endpoint via settings.json.

settings.json (~/.config/zed/settings.json)

{
  "language_models": {
    "openai": {
      "api_url": "https://mecutinai.com/api/v1",
      "available_models": [
        {
          "name": "deepseek-v4-flash",
          "display_name": "DeepSeek V4 Flash",
          "max_tokens": 128000
        }
      ]
    }
  }
}

Set API key via keyring: zed: open keychain → tambah key untuk OpenAI.

🧑‍💻 Aider (CLI Coding Assistant)

Aider support custom OpenAI-compatible endpoint via flag atau config file.

bash

# Via command line flags
aider \
  --openai-api-base "https://mecutinai.com/api/v1" \
  --model deepseek-v4-flash \
  --no-verify-ssl

# Atau via ~/.aider.conf.yml
# openai-api-base: https://mecutinai.com/api/v1
# openai-api-key: sk_AKUN_ANDA
# model: deepseek-v4-flash

🔀 OpenRouter / 9Router

MecutinAI juga bisa dipakai sebagai upstream provider di OpenRouter atau router serupa. Gunakan endpoint OpenAI-compatible standar.

bash

# Environment variables
export OPENAI_BASE_URL="https://mecutinai.com/api/v1"
export OPENAI_API_KEY="sk_AKUN_ANDA"

# Test dengan curl
curl https://mecutinai.com/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

🪿 Hermes / Custom Agent

Untuk agent framework custom (Hermes, dll), set OpenAI client dengan Base URL MecutinAI.

python

from openai import OpenAI

client = OpenAI(
    api_key="sk_AKUN_ANDA",
    base_url="https://mecutinai.com/api/v1",
)

response = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Write a hello world in Python"}],
    stream=True,
)

for chunk in response:
    print(chunk.choices[0].delta.content or "", end="")

Cara Mendapatkan API Key

1Kunjungi halaman /pricing dan pilih plan.
2Klik Beli Sekarang — Anda akan diarahkan ke halaman checkout.
3Scan QRIS dengan e-wallet atau mobile banking. Bayar sesuai nominal (termasuk kode unik).
4Tunggu admin mengonfirmasi pembayaran. Halaman order akan auto-refresh setiap 5 detik.
5Setelah dikonfirmasi, API key muncul di halaman order. Simpan baik-baik — key ditampilkan sekali saja.
6Pakai key tersebut untuk semua request ke /api/v1/*.

Cek sisa kuota: panggil GET /api/v1/usage dengan key Anda, atau lihat header X-RateLimit-Remaining-Tokens di setiap response gateway.

APIDokumentasi

Quick Start

Base URL

Autentikasi

Request Pertama Anda

Model yang Tersedia

Chat Completions

Parameter Body

Contoh — Non-Streaming (curl)

Response (Non-Streaming)

Contoh — Streaming (SSE)

Contoh — JavaScript (fetch)

Contoh — Python (openai SDK)

List Models

Embeddings

Usage

Response (Plan Terbatas)

Response (Plan Unlimited)

Rate Limiting & Kuota

Response Headers

Kode Error

SDK Compatibility

IDE & CLI Integration

🤖 Claude CLI (Claude Code)

🖥 Cursor IDE

📝 Cline (VS Code Extension)

⚡ Continue.dev (VS Code / JetBrains)

🦀 Zed Editor

🧑‍💻 Aider (CLI Coding Assistant)

🔀 OpenRouter / 9Router

🪿 Hermes / Custom Agent

Cara Mendapatkan API Key

API
Dokumentasi