Docs

Overview and quickstart

RouterLane is an LLM API where you pick a tier instead of a model. Send fast, normal, smart or smart-plus as the model name, and the router picks the model and reasoning effort for each task.

It speaks three API formats, so the SDK or agent you already use works by changing its base URL and key.

Format	Endpoint	Base URL for SDKs
Anthropic Messages	`POST /v1/messages`	`https://api.routerlane.com`
OpenAI Chat Completions	`POST /v1/chat/completions`	`https://api.routerlane.com/v1`
OpenAI Responses	`POST /v1/responses`	`https://api.routerlane.com/v1`

1. Store your key

Keys are issued on request: request access and the key arrives by email. It starts with rl_. Put it in an environment variable named ROUTERLANE_API_KEY; every example in these docs reads it from there.

2. Call it in the OpenAI format

shell

curl https://api.routerlane.com/v1/chat/completions \
  -H "Authorization: Bearer $ROUTERLANE_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "normal",
    "messages": [{"role": "user", "content": "Write a regex that matches dates like 2026-10-06."}]
  }'

quickstart.py

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.routerlane.com/v1",
    api_key=os.environ["ROUTERLANE_API_KEY"],
)

response = client.chat.completions.create(
    model="normal",
    messages=[{"role": "user", "content": "Write a regex that matches dates like 2026-10-06."}],
)
print(response.choices[0].message.content)

quickstart.ts

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.routerlane.com/v1",
  apiKey: process.env.ROUTERLANE_API_KEY,
});

const response = await client.chat.completions.create({
  model: "normal",
  messages: [{ role: "user", content: "Write a regex that matches dates like 2026-10-06." }],
});
console.log(response.choices[0].message.content);

3. Or in the Anthropic format

shell

curl https://api.routerlane.com/v1/messages \
  -H "x-api-key: $ROUTERLANE_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "smart",
    "max_tokens": 4096,
    "messages": [{"role": "user", "content": "Review this function for race conditions."}]
  }'

quickstart.py

import os
import anthropic

client = anthropic.Anthropic(
    base_url="https://api.routerlane.com",
    api_key=os.environ["ROUTERLANE_API_KEY"],
)

message = client.messages.create(
    model="smart",
    max_tokens=4096,
    messages=[{"role": "user", "content": "Review this function for race conditions."}],
)
print("".join(block.text for block in message.content if block.type == "text"))

quickstart.ts

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  baseURL: "https://api.routerlane.com",
  apiKey: process.env.ROUTERLANE_API_KEY,
});

const message = await client.messages.create({
  model: "smart",
  max_tokens: 4096,
  messages: [{ role: "user", content: "Review this function for race conditions." }],
});
for (const block of message.content) {
  if (block.type === "text") console.log(block.text);
}

4. Read the decision

Every response says what the router did. The model field in the body holds the tier you called; the headers name the model and effort behind it.

response headers

HTTP/2 200
content-type: application/json
x-router-tier: smart
x-router-model: glm-5.3
x-router-effort: high
x-router-source: rule
x-router-request-id: 4f1c9a07d2e86b13

x-router-source tells you how the decision was made: rule or default for a new task, continuation for a tool result inside a running task, sticky when the conversation kept its model, and smartplus for a Smart+ job turn. See response headers.

Overview and quickstart

1. Store your key

2. Call it in the OpenAI format

3. Or in the Anthropic format

4. Read the decision

Next

Integrations

Tiers and model IDs

API reference

Errors and limits