Installation & Setup

The RunRL Python SDK (runrl) provides a typed client for the RunRL REST API. This page covers installation, authentication, and a minimal smoke test.

1. Install the SDK

pip install runrl==0.2.1

The package supports Python 3.9+ and ships with both synchronous and asynchronous clients. It does not install heavy ML dependencies—only standard HTTP tooling (httpx + pydantic).

2. Configure Credentials

Generate an API key from the RunRL dashboard and expose it to your environment. The client reads the RUNRL_API_KEY variable automatically; you can also pass it explicitly.

export RUNRL_API_KEY="rl_your_api_key"

# Optional: target staging or local environments
export RUNRL_BASE_URL="https://staging.runrl.com/api/v1"

3. Verify Connectivity

from runrl import RunRLClient

client = RunRLClient()
print(client.config.base_url)

# list your files to ensure auth works
files = client.files.list(per_page=5)
for file in files.items:
    print(file.id, file.name, file.type)

If this returns a AuthenticationError, double-check the API key. Network or SSL errors typically indicate proxies/firewalls blocking the request.

4. Optional: Async Client

import asyncio
from runrl import AsyncRunRLClient


async def main():
    async with AsyncRunRLClient() as client:
        files = await client.files.list(per_page=3)
        print([f.name for f in files.items])

asyncio.run(main())

5. Local Development Helpers

• User agent: Overwrite via RunRLClient(user_agent="my-app/0.1.0").
• Timeouts: RunRLClient(timeout=60) adjusts HTTP timeouts.
• Polling intervals: RunRLClient(poll_interval=10) tunes future polling frequency.

With the SDK installed and authenticated, continue to Launch Your First Run or jump straight to the API Reference.