Switch from OpenAI API to RouteScope

Minimal changes for moving an existing OpenAI SDK, curl request, or OpenAI-compatible client to RouteScope.

If your project already calls the OpenAI API, switching to RouteScope usually means changing three values: Base URL, API Key, and model name. The request format can stay OpenAI-compatible.

Setting	OpenAI	RouteScope
Base URL	`https://api.openai.com/v1`	`https://api.routescope.ai/v1`
API Key	OpenAI key	RouteScope API Key starting with `sk-...`
Model	OpenAI model name	A model copied from Model Plaza or `/v1/models`

Use the /v1 Base URL

OpenAI-compatible clients should use https://api.routescope.ai/v1. Do not use the website URL, and do not omit /v1.

Step 1: Create a RouteScope API Key

Create an API Key in the RouteScope console and copy the full sk-... value.

If you are not sure which models your account can call, check Model Plaza or call:

curl https://api.routescope.ai/v1/models \
  -H "Authorization: Bearer sk-your-token"

Step 2: Update your OpenAI SDK settings

JavaScript / TypeScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.ROUTESCOPE_API_KEY,
  baseURL: "https://api.routescope.ai/v1",
});

const completion = await client.chat.completions.create({
  model: "your-routescope-model",
  messages: [{ role: "user", content: "Reply with RouteScope connected" }],
});

console.log(completion.choices[0]?.message?.content);

If your application already reads OPENAI_API_KEY, you can keep that environment variable name. Set its value to the RouteScope API Key and change the Base URL.

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-token",
    base_url="https://api.routescope.ai/v1",
)

resp = client.chat.completions.create(
    model="your-routescope-model",
    messages=[{"role": "user", "content": "Reply with RouteScope connected"}],
)

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

curl

curl https://api.routescope.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-token" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-routescope-model",
    "messages": [
      {
        "role": "user",
        "content": "Reply with RouteScope connected"
      }
    ]
  }'

Step 3: Keep the OpenAI-compatible paths

Capability	RouteScope path
Model list	`GET /v1/models`
Chat completions	`POST /v1/chat/completions`
Image generation	`POST /v1/images/generations`
Image edits	`POST /v1/images/edits`

For example, chat completions use https://api.routescope.ai/v1/chat/completions.

Step 4: Verify the request

After a successful request, open operation records and check:

Item	What to verify
Model	It matches the RouteScope model you requested.
Token	It matches the API Key you created.
Status	The request succeeded, or the failure reason is clear.
Usage	Token usage and cost were recorded.

Common issues

Symptom	Fix
`401` or `403`	Check that the API Key is complete, enabled, and has quota and permission.
`404` or model not found	Copy a model name from Model Plaza or `/v1/models`.
Requests still go to OpenAI	Check whether an old Base URL remains in the SDK, proxy, or client.
Claude-native calls fail	Claude Messages uses `https://api.routescope.ai/v1/messages`; do not send it through an OpenAI SDK path.