Quickstart

Send the same chat completion request using the stack you prefer. All examples target the OpenAI-compatible gateway at https://router.hpp.io.

Prerequisites

• An HPP Router API key from HPP Hub
• Base URL: https://router.hpp.io

Store your key in an environment variable:

export HPPROUTER_API_KEY="your-api-key"

Send a chat completion

Install the package you need before running an SDK example:

• TypeScript SDK: npm install @hpprouter/sdk
• OpenAI SDK: npm install openai
• Python: pip install openai

TypeScript SDK

import { HppRouter } from '@hpprouter/sdk';

const client = new HppRouter({
  apiKey: process.env.HPPROUTER_API_KEY!,
  baseURL: 'https://router.hpp.io',
});

const completion = await client.chat.send({
  model: 'hpprouter/auto',
  messages: [{ role: 'user', content: 'Hello!' }],
});

console.log(completion.data.choices);
console.log(completion.meta.resolvedModel);

OpenAI SDK

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.HPPROUTER_API_KEY!,
  baseURL: 'https://router.hpp.io/llm/v1',
});

const completion = await client.chat.completions.create({
  model: 'hpprouter/auto',
  messages: [{ role: 'user', content: 'Hello!' }],
});

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

cURL

curl -X POST https://router.hpp.io/llm/v1/chat/completions \
  -H "apikey: $HPPROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hpprouter/auto",
    "messages": [
      { "role": "user", "content": "Hello!" }
    ],
    "max_completion_tokens": 100
  }'

Python

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["HPPROUTER_API_KEY"],
    base_url="https://router.hpp.io/llm/v1",
)

completion = client.chat.completions.create(
    model="hpprouter/auto",
    messages=[{"role": "user", "content": "Hello!"}],
)

print(completion.choices[0].message)

See the Client SDK and OpenAI SDK (drop-in) guides for more examples.

Response

A successful response includes choices and a usage block:

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "openai/gpt-4o-mini",
  "choices": [
    { "message": { "role": "assistant", "content": "Hi there!" } }
  ],
  "usage": {
    "prompt_tokens": 8,
    "completion_tokens": 12,
    "total_tokens": 20
  }
}

When you use hpprouter/auto, the model used for billing is returned in the X-HPP-Router-Resolved-Model response header. See Smart Routing.

Next steps

• Authentication — API key vs. Bearer token.
• Models & Pricing — discover available models.
• Guides — chat, streaming, vision, images, quota, and errors.