Rsby

Unified LLM gateway

One API for every AI model. Swap models, not your code.

Call OpenAI, Claude, and Gemini through one OpenAI-compatible endpoint. Pay only for what you use, with no subscriptions and no lock-in.

Get started freeRead the docs

Free to try. No credit card required.

Drop-in compatible with the OpenAI SDK

Just swap the base URL. Existing OpenAI code keeps working — no rewrite needed.

import OpenAI from 'openai'

const client = new OpenAI({
  apiKey: process.env.RSBY_API_KEY,    // sk-rsby-...
  baseURL: 'https://api.rsby.dev/v1',  // <- satu-satunya perubahan
})

const res = await client.chat.completions.create({
  model: 'gpt-4o-mini',
  messages: [{ role: 'user', content: 'Halo, apa kabar?' }],
})

console.log(res.choices[0].message.content)

Responses follow the OpenAI spec. Streaming, function calling, and multi-provider models are supported.

Supported providers

Add a new provider — your code stays the same.

  • OpenAIAvailable
  • AnthropicSoon
  • GoogleSoon

Why Rsby?

Built for developers who want one simple, fairly priced way to reach every model.

Pay your way

Top up with a card or the local methods you already use — QRIS, virtual account, and e-wallet in Indonesia, with more regions on the way. No hidden conversion fees.

One API, many models

Switch from GPT-4 to Claude or Gemini with a single parameter. No multiple SDKs to learn.

Transparent pricing

A thin margin on top of provider rates. See remaining credit and per-request cost live on your dashboard.

Full control

Create and revoke API keys anytime. Monitor usage, credit limits, and per-request logs — all from one place.

Frequently asked

  • Do I need a credit card to sign up?

    No. Sign up free with email or Google. A credit card is just one payment option — local methods like QRIS, virtual account, and e-wallet are available in supported regions.

  • Can I keep using my existing code?

    Yes, if you're using the OpenAI SDK. Change the base URL to Rsby and use your Rsby API key. No other code changes required.

  • Are my prompts stored?

    We store usage metadata (token counts, cost, model) for billing and analytics. Prompt content and responses are not stored by default.

  • How is each request priced?

    Cost = the provider rate for that request + a thin margin, billed in USD. Local currencies (such as IDR) are converted at the daily rate. Full details on the Pricing page after sign-in.

  • What happens when my credit runs out mid-request?

    The request is rejected with a clear error before reaching the provider. You will not be charged beyond your credit balance.