Becoming a Provider

Becoming a Provider

Want to earn by running jobs on your machine? Become an Auralis provider.

What is a Provider?

Providers are people who share their compute power with the Auralis network. When someone submits a job, your machine can pick it up, run it, and earn credits.

All you need is:

  • A computer with Docker installed
  • A stable internet connection
  • Python 3.9 or newer

Getting Started

Step 1: Get Your Worker Token

Head to Dashboard → Provide. We'll automatically detect your system's platform (AMD64 or ARM64).

Click Register as Worker to get your unique token. Save this somewhere safe – you'll only see it once!

Step 2: Install the Worker CLI

Open your terminal and install the Auralis worker:

pip install auralis-worker

That's it. One command.

Step 3: Start Running Jobs

There are two ways to run jobs:

Option A: Run a specific job

See a job you want to run? Copy its ID from the dashboard and:

auralis-worker --token YOUR_TOKEN --job-id JOB_ID

Option B: Auto-pick jobs (coming soon)

Let the worker automatically pick up jobs:

auralis-worker --token YOUR_TOKEN

What Happens When You Run a Job

  1. Download – The worker downloads the project files
  2. Build – Docker builds the image from the Dockerfile
  3. Run – The container runs with the specified timeout
  4. Upload – Output files are uploaded back to Auralis
  5. Complete – Job status updates and you earn credits

Here's what it looks like in action:

    +=======================================+
    |     AURALIS CLOUD WORKER NODE         |
    |   Distributed AI Compute Platform     |
    +=======================================+

2025-12-14 09:18:31 - INFO - [OK] Authenticated as: My Worker
2025-12-14 09:18:35 - INFO - [DOWNLOAD] Downloading input...
2025-12-14 09:18:36 - INFO - [BUILD] Building Docker image
2025-12-14 09:19:12 - INFO - [RUN] Running container (1h timeout)...
2025-12-14 09:20:22 - INFO - [UPLOAD] Uploading outputs...
2025-12-14 09:20:25 - INFO - [OK] Job completed successfully

Requirements

Docker

Docker must be running on your machine. Verify with:

docker info

If you get an error, start Docker Desktop or the Docker daemon.

Platform Matching

Your worker only sees jobs that match your platform:

  • AMD64 worker → only sees AMD64 jobs
  • ARM64 worker → only sees ARM64 jobs

This prevents architecture mismatches and failed jobs.

Earnings (Coming Soon)

The credits and earnings system is under development. Soon you'll be able to:

  • Earn credits for each completed job
  • Withdraw earnings to your account
  • Track your provider stats

Troubleshooting

"Failed to claim job"

The job was claimed by another worker first. Try another job or run in polling mode.

"Docker build failed"

The uploaded project has issues. This isn't your fault – the job submitter needs to fix their Dockerfile.

"No output files to upload"

The job completed but didn't write anything to the output directory. The submitter needs to save files to ./output.

Submitting JobsAPI Reference