SDK

Quickstart

Get started with Hyperion in minutes. Our SDKs provide a drop-in replacement for OpenAI-compatible clients with built-in caching and smart routing.

01 — INSTALLATION

Add to Project

Install the official Hyperion library using your preferred package manager (pip, npm, or pnpm).

02 — CONFIGURATION

Initialize Client

Pass your Hyperion API Key and the Gateway endpoint to the client constructor.

03 — EXECUTION

Chat Completion

Use the familiar chat completions interface. Hyperion handles the provider routing automatically.

04 — OPTIMIZATION

Check Headers

Inspect the 'X-Cache' and 'X-Gateway-Dispatch' headers to verify performance gains.

Basic Usage

Switch between languages to see how to initialize the Hyperion client and make your first proxied request.

from hyperion import HyperionClient

client = HyperionClient(
    api_key="your_hyperion_key",
    base_url="https://api.hyperion.ai/v1"
)

# Call any model via the gateway
response = client.chat.completions.create(
    model="gpt-4.1-nano",
    messages=[{"role": "user", "content": "Hello!"}]
)

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

Installation

Python (pip)

pip install hyperion-ai

TypeScript (npm)

npm install @hyperion-ai/sdk