Do I need to replace the OpenAI library I already use?

No. Keep the same library and change the base URL, API key, and model. The Python example on this page uses the OpenAI package without an extra adapter.

Which models are available on the chat endpoint?

The current catalog includes GPT, Claude, Gemini, Grok, DeepSeek, and Kimi families. GET /v1/models is the current source for model IDs and availability.

Are streaming and tool calling supported?

The chat endpoint supports streaming. Tool calling depends on the model and is listed in that model’s catalog capabilities.

Is the bill denominated in dollars?

No. Token prices and the account balance are shown in rupiah, and the balance is topped up through QRIS.

GetRun / OpenAI-compatible API

An OpenAI-Compatible API for Developers in Indonesia

An application already using the OpenAI SDK can move to GetRun by changing its base URL, API key, and model value. POST /v1/chat/completions preserves the messages format, SSE streaming, tool calling on models that support it, and an OpenAI-compatible response shape. GPT, Claude, Gemini, Grok, DeepSeek, and Kimi can be selected without rebuilding the integration for each provider.

This page draws on 22 models from 6 providers in the GetRun catalog, with sample runs starting at Rp495 each. All of them are paid from one rupiah balance, topped up through QRIS.

Why through GetRun

One base URL, https://getrun.id/v1, for an OpenAI SDK integration you already use.
Switch among GPT, Claude, Gemini, Grok, DeepSeek, and Kimi by changing only the model value.
SSE streaming returns chat tokens as they are generated.
Tool calling is available on models that list that capability in the catalog.
Input, cached-input, and output tokens are billed in rupiah.
One balance and one usage report covers every provider.

The steps

Change the base URL — Point the OpenAI SDK at https://getrun.id/v1 and use a GetRun API key as the bearer token.
Choose a model — Take an ID from GET /v1/models and put it in the model field without changing your existing messages format.
Enable streaming when needed — Send stream: true to receive server-sent events in an OpenAI-compatible chunk format.

Example with the OpenAI SDK

Point base_url at GetRun and use a model ID from GET /v1/models. This example uses one of the chat models currently in the catalog.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_GETRUN_API_KEY",
    base_url="https://getrun.id/v1",
)

response = client.chat.completions.create(
    model="gpt/6-luna",
    messages=[{"role": "user", "content": "Hello from Indonesia"}],
)

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

Available models and pricing

AI Text & Chat Models (22)

Model	Provider	Price
GPT-6 Luna	OpenAI	Rp495 / 1M token input
DeepSeek V4.1 Flash	DeepSeek	Rp1,485 / 1M token input
Gemini 3.8 Flash	Google	Rp2,784 / 1M token input
GPT-5.4 Mini	OpenAI	Rp3,713 / 1M token input
GPT-5.6 Luna	OpenAI	Rp4,950 / 1M token input
GPT-5.3 Codex Spark	OpenAI	Rp8,663 / 1M token input
GPT-6.1 Sol	OpenAI	Rp9,900 / 1M token input
GPT-6 Sol	OpenAI	Rp9,900 / 1M token input
Claude Sonnet 5.5	Anthropic	Rp9,900 / 1M token input
Grok 4.7	xAI	Rp9,900 / 1M token input
Grok 4.5	xAI	Rp9,900 / 1M token input
Grok 4.6	xAI	Rp9,900 / 1M token input
Claude Sonnet 5	Anthropic	Rp10,519 / 1M token input
GPT-5.6 Terra	OpenAI	Rp12,375 / 1M token input
GPT-5.4	OpenAI	Rp12,375 / 1M token input
Claude Opus 5.5	Anthropic	Rp19,800 / 1M token input
GPT-5.6 Sol	OpenAI	Rp24,750 / 1M token input
GPT-5.5	OpenAI	Rp24,750 / 1M token input
Claude Opus 4.8	Anthropic	Rp24,750 / 1M token input
Kimi K3	Moonshot AI	Rp29,700 / 1M token input
GPT-6 Astra	OpenAI	Rp49,500 / 1M token input
Claude Fable 5	Anthropic	Rp49,500 / 1M token input

Frequently asked questions

Do I need to replace the OpenAI library I already use?: No. Keep the same library and change the base URL, API key, and model. The Python example on this page uses the OpenAI package without an extra adapter.
Which models are available on the chat endpoint?: The current catalog includes GPT, Claude, Gemini, Grok, DeepSeek, and Kimi families. GET /v1/models is the current source for model IDs and availability.
Are streaming and tool calling supported?: The chat endpoint supports streaming. Tool calling depends on the model and is listed in that model’s catalog capabilities.
Is the bill denominated in dollars?: No. Token prices and the account balance are shown in rupiah, and the balance is topped up through QRIS.

Related guides

Standards we follow

Katalog lengkap · Daftar harga · Bandingkan model · Dokumentasi API · Tentang GetRun