API relay station for multi-model AI access

Build your own API relay workflow: prepay credits upstream, create relay keys, point downstream apps at one OpenAI-compatible endpoint, and monitor spend and stability.

Base URL
https://api.duang-duang.com/v1

POST /chat/completions
Authorization: Bearer sk-relay_...

{
  "model": "budget",
  "messages": [
    {"role": "user", "content": "Summarize this document"}
  ]
}

Agency or relay? Duang Duang is built for relay mode.

Reseller mode

Good for selling a service package, but pricing power and product control usually stay with the upstream provider.

API relay mode

Buy upstream capacity first, expose your own relay endpoint, serve downstream apps, and own the usage experience.

Responsibility built in

Credits, keys, usage charts, model pricing, and request status are visible so you can handle balance, no-response, and support issues.

How it works

1

Create an account

Sign up in the dashboard and verify your email before using credits or API keys.

2

Add prepaid credits

Buy a preset credit package or enter a custom amount. Stripe handles payment securely.

3

Create your API key

Generate a key that starts with sk-relay_. Copy it when it is shown.

4

Use the API

Set the base URL to https://api.duang-duang.com/v1 in compatible tools or apps.

What we offer

Unified API

Use one API interface and one customer key to access supported model providers through a managed gateway.

Prepaid credits

Buy credits before use, avoid surprise invoices, and monitor balance and token usage in real time.

Model routing

Choose budget, standard, or premium tiers depending on cost, speed, and quality needs.

Pricing

Budget

Lower-cost routing for high-volume text workloads.

$0.12 input / $0.48 output per 1M tokens

Standard

Balanced routing for everyday application workloads.

$0.36 input / $3.00 output per 1M tokens

Premium

Higher-capability models for more complex reasoning and writing tasks.

$3.60 input / $18.00 output per 1M tokens

Credits are prepaid in USD cents. Usage is deducted based on selected model, input tokens, and output tokens. Tiers route to supported provider models and are priced as Duang Duang customer rates.

API setup

Use in OpenAI-compatible tools

Use your Duang Duang key as the API key and set the base URL to our endpoint.

OPENAI_API_KEY=sk-relay_YOUR_KEY
OPENAI_BASE_URL=https://api.duang-duang.com/v1

Common model names

  • budget - low-cost default tier
  • standard - balanced tier
  • premium - higher-capability tier
  • openai/gpt-5.4-mini
  • anthropic/claude-sonnet-4.6
  • google/gemini-2.5-flash
curl https://api.duang-duang.com/v1/chat/completions \
  -H "Authorization: Bearer sk-relay_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "budget",
    "messages": [
      {"role": "user", "content": "Say ok in one word."}
    ],
    "max_tokens": 50
  }'

Compatibility depends on whether your app supports a custom OpenAI-compatible base URL.

Compliance and support

Usage controls

Accounts have credit balances, API keys, rate limits, and usage logs.

Policies

Customers must follow our Terms, Privacy Policy, Refund Policy, Acceptable Use Policy, and upstream provider rules.

Contact

Support email: support@duang-duang.com.

Terms of Service

Duang Duang provides prepaid API credits and a software gateway for accessing supported AI model providers through a unified API. Customers are responsible for keeping API keys secure and for all usage made through their keys.

Credits are prepaid and deducted based on model usage, including input tokens, output tokens, and applicable pricing rules. Customers must follow applicable upstream provider policies and legal requirements. Duang Duang does not grant permission to bypass provider restrictions, rate limits, or safety policies.

Privacy Policy

We collect account information, API key metadata, payment records, usage records, and technical logs needed to operate the service. We record token usage, selected models, request status, provider cost, billed amount, and timestamps.

Payment processing is handled by Stripe. We do not store full card numbers on our servers. We may share necessary request data with upstream AI model providers to fulfill API requests.

Refund Policy

Prepaid credits are generally refundable only for unused balances, billing errors, duplicate charges, or service issues reviewed by support. Credits consumed by API usage are not refundable unless the usage was caused by a confirmed platform error.

Refund requests should include the account email, payment date, amount, and reason. Contact support@duang-duang.com.

Acceptable Use Policy

Customers must use Duang Duang lawfully and in compliance with upstream AI provider policies. Prohibited uses include illegal activity, fraud, credential theft, phishing, malware, harassment, attempts to bypass provider restrictions, and automated activity that harms service stability or violates rate limits.

We may throttle, suspend, or terminate accounts and API keys that violate this policy. Report abuse to support@duang-duang.com.