Anthropic and OpenAI models, one gateway

Claude and GPT models,on one API key

Buy a plan for 1, 3, 7 or 30 days. You get a base URL and a key, and every request is unlimited for the whole period. Point Qoder, VS Code, Cursor, Claude Code or a plain terminal at it and the models work.

Buy 1 day of access, Rp 30.000 See the setup for your editor

Basic models: 5
Extension models: 3
Context per model: 1M tokens
Upstream: Reachable

What you get

Upstream reachable

Base URL (OpenAI-compatible)

https://clauderouter.my.id/v1

Base URL (Anthropic Messages API)

https://clauderouter.my.id/v1

Your API key

••-••••••••••••••••••••••••••••••••

The same key works for both formats. Nothing else in your client needs to change.

Currently served by the upstream, checked 02:41:37

anthropic/claude-sonnet-4.6served
anthropic/claude-opus-4.7served
anthropic/claude-opus-4.8served
anthropic/claude-opus-5served
openai/gpt-5.6-solserved

Full status for every model

What is in the catalog

Two groups. Basic models come with any plan. Extension models need an active basic key first, because they attach to a key you already own.

Basic tier

Included in every plan

Claude Sonnet 4.6anthropic/claude-sonnet-4.61,000,000 ctx, vision
Claude Opus 4.7anthropic/claude-opus-4.71,000,000 ctx, vision
Claude Opus 4.8anthropic/claude-opus-4.81,000,000 ctx, vision
Claude Opus 5anthropic/claude-opus-51,000,000 ctx, vision
GPT 5.6 Solopenai/gpt-5.6-sol1,000,000 ctx, vision

Extension tier

Needs a basic key

Claude Fable 5anthropic/claude-fable-5
Extended-reasoning work in the extension tier
Claude Fable 5.1anthropic/claude-fable-5.1
The most demanding Claude work in this catalog
GPT 6 Astraopenai/gpt-6-astra
Frontier reasoning for tasks the basic tier does not cover

From Rp 10.000 for 1 day, added to a key you already have.

The gateway is the product

It speaks the OpenAI Chat Completions format and the Anthropic Messages format. Any client that already understands one of those works without a plugin, a proxy setting on your machine, or a local shim.

Unlimited for the length of the plan
No token meter and no request cap while the plan is running. The limit is the clock, not a quota.
Expiry is exact
When the window closes, the key stops answering. There is no grace period and nothing to cancel.
Vision input
Every model in the catalog accepts images, so screenshots and diagrams work in your editor.
1M token context
Large files and long agent runs fit without trimming, on every model listed.

Request

curl https://clauderouter.my.id/v1/chat/completions \
  -H "Authorization: Bearer $CLAUDEROUTER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.8",
    "messages": [
      { "role": "user", "content": "Explain this stack trace." }
    ]
  }'

Response

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "anthropic/claude-opus-4.8",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "..." },
    "finish_reason": "stop"
  }]
}

The Anthropic format is served at the same base URL under the same key, so Claude Code and the Anthropic SDKs talk to it unchanged.

Basic first, then extension

The extension tier is an add-on, not a separate product. It attaches to a key you already own, which is why the order below is the only one that works.

1
Buy a basic plan
Choose 1 day, 3 days, 7 days, 30 days. You get a key immediately, carrying every basic model.
2
Point your editor at it
Set the base URL and paste the key. Both API formats are live on the same endpoint.
3
Add the extension when you want it
Attach 1 day, 3 days, 7 days, 30 days of extension access to that same key. Basic has to still be active.

Where it runs

Any client that lets you set an API base URL. The full setup for each is in the documentation, with the exact values to paste.

Qoder
Settings, Models, add a custom OpenAI-compatible provider
VS Code
Continue, Cline, Roo Code and any extension with a base URL field
Cursor
Custom OpenAI base URL override
Claude Code
ANTHROPIC_BASE_URL and ANTHROPIC_AUTH_TOKEN
Terminal
curl, plus the OpenAI and Anthropic Python and Node SDKs
Any OpenAI-compatible client
LibreChat, Open WebUI, SillyTavern, LangChain

Open the setup documentation

Prices

Paid in rupiah through QRIS or a bank virtual account. Access starts as soon as the payment confirms.

Basic plans

Every basic model, unlimited requests.

Basic plan prices
Length	Price
1 day	Rp 30.000
3 days	Rp 50.000
7 days	Rp 100.000
30 days	Rp 300.000

Buy a basic plan

Extension plans

Requires an active basic key. Adds the extension models to that key.

Extension plan prices
Length	Price
1 day	Rp 10.000
3 days	Rp 15.000
7 days	Rp 25.000
30 days	Rp 50.000

Add an extension

4 accounts created, 6 active keys, 192 requests in the last 24 hours.

Create an account, buy a plan, and copy the key. The whole setup takes about a minute.

Create an account Browse the catalog