Setup

Qoder

Qoder accepts a custom OpenAI-compatible provider. Add one and point it here.

1. 1Open Qoder settings and go to the model or provider section.
2. 2Add a new custom provider, choosing the OpenAI-compatible type.
3. 3Set the base URL to https://clauderouter.my.id/v1
4. 4Paste your key as the API key.
5. 5Add the models you want by id, for example anthropic/claude-sonnet-4.6 and anthropic/claude-opus-5.
6. 6Select the provider as active.

The model list endpoint is live, so a client that reads it will populate its picker with the catalog automatically:

curl https://clauderouter.my.id/v1/models \
  -H "Authorization: Bearer cr-your-key-here"

VS Code extensions

Continue, Cline, Roo Code and similar extensions all take an OpenAI-compatible base URL. The field names differ slightly, the values do not.

// Continue, config.yaml
models:
  - name: Claude Opus 4.8
    provider: openai
    model: anthropic/claude-opus-4.8
    apiBase: https://clauderouter.my.id/v1
    apiKey: cr-your-key-here

// Cline / Roo Code, settings
API Provider:  OpenAI Compatible
Base URL:      https://clauderouter.my.id/v1
API Key:       cr-your-key-here
Model ID:      anthropic/claude-sonnet-4.6

Extensions that ask for the model name rather than the model id should be given the id exactly as written, including the provider prefix.

Cursor

In Cursor, open Settings, then Models, and add an OpenAI-compatible entry with an overridden base URL.

Base URL:  https://clauderouter.my.id/v1
API Key:   cr-your-key-here
Model:     anthropic/claude-opus-4.8

Turn off any other providers for the model you add, since Cursor will otherwise keep routing that model to its own backend.

Terminal

A plain request. Both formats are shown so you can confirm the key works before configuring an editor.

# OpenAI-compatible
curl https://clauderouter.my.id/v1/chat/completions \
  -H "Authorization: Bearer cr-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.8",
    "messages": [{"role": "user", "content": "Reply with the single word: ready"}]
  }'

# Anthropic Messages format, same key
curl https://clauderouter.my.id/v1/messages \
  -H "x-api-key: cr-your-key-here" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.8",
    "max_tokens": 64,
    "messages": [{"role": "user", "content": "Reply with the single word: ready"}]
  }'

OpenAI SDK

The official OpenAI SDK works by changing only the base URL. This is also the path for LangChain, LlamaIndex and anything else built on it.

from openai import OpenAI

client = OpenAI(
    base_url="https://clauderouter.my.id/v1",
    api_key="cr-your-key-here",
)

stream = client.chat.completions.create(
    model="anthropic/claude-opus-4.8",
    messages=[{"role": "user", "content": "Hello"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://clauderouter.my.id/v1",
  apiKey: "cr-your-key-here",
});

const res = await client.chat.completions.create({
  model: "anthropic/claude-opus-4.8",
  messages: [{ role: "user", content: "Hello" }],
});
console.log(res.choices[0].message.content);

Anthropic SDK

The Anthropic SDKs work as well. Note that the base URL has no trailing/v1, because the SDK adds it.

from anthropic import Anthropic

client = Anthropic(
    base_url="https://clauderouter.my.id",
    api_key="cr-your-key-here",
)

message = client.messages.create(
    model="anthropic/claude-opus-4.8",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
print(message.content[0].text)

Other OpenAI-compatible apps

LibreChat, Open WebUI, SillyTavern, Jan, Chatbox and similar applications take a base URL and a key in their own settings screens. The values are always: