One OpenAI-compatible API, every ModelRouters model.

Connect a chat client

Choose Chat Completion and a custom OpenAI-compatible source. Set the Base URL to https://api.modelrouters.net/v1, enter your ModelRouters API key, and use a hosted model id, an original OpenRouter model id, or one of your optional private user/<alias> model ids. When the client appends /chat/completions automatically, do not add that suffix to the Base URL yourself.

Connect a provider

Bring-your-own-key routing is free. DeepSeek is the recommended direct BYOK starting point, while other supported providers follow the same setup. Your provider bills model usage directly; ModelRouters keeps the connection credential encrypted at rest.

1. 1. Add a provider. Choose the provider, paste its API key, and save the connection.
2. 2. Test the connection. Confirm the credential and upstream endpoint work before routing chat.
3. 3. Route OpenRouter directly. With an OpenRouter connection, send its original provider/model id unchanged.
4. 4. Add aliases only if useful. Optional user/<alias> profiles add a friendly name, preset, persona, or roleplay mode.

Quick start with curl

Create a key in the dashboard, keep it server-side, and send a normal Chat Completions request.

export MODELROUTERS_KEY="mr_live_..."

curl  https://api.modelrouters.net/v1/chat/completions \
  -H "Authorization: Bearer $MODELROUTERS_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [
      {"role": "user", "content": "Write a two-line welcome message."}
    ]
  }'

Use the OpenAI SDK

Existing OpenAI Chat Completions integrations usually need only a new base URL, API key, and model id.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["MODELROUTERS_KEY"],
    base_url=" https://api.modelrouters.net/v1",
)

response = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.MODELROUTERS_KEY,
  baseURL: " https://api.modelrouters.net/v1",
});

const response = await client.chat.completions.create({
  model: "deepseek-v4-flash",
  messages: [{ role: "user", content: "Hello" }],
});
console.log(response.choices[0].message.content);

Streaming

Set stream: true. ModelRouters returns standard server-sent Chat Completions chunks.

const stream = await client.chat.completions.create({
  model: "claude-sonnet-4.6",
  messages: [{ role: "user", content: "Tell me a short story." }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

Endpoints and responses