Hello, world!

A complete, runnable first script with Radium

Hello, world!

This is a complete, minimal Radium application. Copy it into a file, set your API key, and run it.

The script

hello.py

from openai import OpenAI
import os

# 1. Create a client — just change base_url and api_key
client = OpenAI(
    api_key=os.environ["RADIUM_API_KEY"],
    base_url="https://api.radium.cloud/v1",
)

# 2. Send a chat completion request
response = client.chat.completions.create(
    model="clarke-1.0",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "What is quantum computing in one sentence?"},
    ],
    temperature=0.7,
    max_tokens=256,
)

# 3. Print the response
print(response.choices[0].message.content)

# 4. (Optional) Inspect token usage
print(f"Tokens used: {response.usage.total_tokens}")

Run it

export RADIUM_API_KEY="YOUR_RADIUM_API_KEY"
python hello.py

What it does

1. Creates an OpenAI-compatible client pointing at api.radium.cloud/v1.
2. Sends a chat request to clarke-1.0 (our Sonnet-class model, ideal for most tasks).
3. Receives and prints the model’s text response.
4. Logs usage so you can see how many tokens were consumed.

Switch to streaming

Replace the non-streaming call with this to get tokens as they arrive:

stream = client.chat.completions.create(
    model="clarke-1.0",
    messages=[{"role": "user", "content": "Count to 10."}],
    stream=True,
)

for chunk in stream:
    delta = chunk.choices[0].delta.content or ""
    print(delta, end="", flush=True)
print()

Switch to Anthropic format

Same logic, different SDK and base URL:

hello_anthropic.py

from anthropic import Anthropic
import os

client = Anthropic(
    api_key=os.environ["RADIUM_API_KEY"],
    base_url="https://api.radium.cloud",
)

response = client.messages.create(
    model="clarke-1.0",
    max_tokens=256,
    messages=[{"role": "user", "content": "What is quantum computing in one sentence?"}],
)

print(response.content[-1].text)

Next steps

Streaming response

Stream tokens in real time

Tool calling agent

Build an agent that uses tools

Structured extraction

Extract JSON from text

Models overview

Pick the right model tier