← Back to Documentation

Operator Earnings Guide

How much you earn, what affects your rate, and how payouts work.

The Short Version

Simulacrum uses flat-rate pricing. Every job pays the same rate per GPU-hour.

You earn $0.60/hr (80% of the $0.75 client rate)

What You Earn Per Hour

Rate
Client pays $0.75/hr
Your cut (80%) $0.60/hr
Platform fee (20%) $0.15/hr

All billing is per-second. You are paid for the exact GPU time consumed, not rounded to the hour.

What Runs on Your Node

Every job runs in a hardened Docker container with capability dropping, resource isolation, and GPU passthrough. Project files are stored in a temp directory during execution and securely wiped after the job completes.

Electricity Costs

Electricity cost depends only on your GPU power draw.

GPU Power Draw Cost/hr (@ $0.12/kWh) Net Profit/hr
RTX 3060170W$0.020$0.58
RTX 3080320W$0.038$0.56
RTX 4070200W$0.024$0.58
RTX 4090450W$0.054$0.55
RTX 5090575W$0.069$0.53

Net profit at the flat $0.60/hr operator rate varies by GPU power draw. Lower-wattage GPUs are more profitable.

Monthly Earnings Scenarios

Scenario Earnings ($0.60/hr)
Light — 2 hrs/day, 20 days $24
Moderate — 8 hrs/day, 25 days $120
Heavy — 20 hrs/day, 30 days $360

Actual earnings depend on job availability. Your node earns money only during active compute jobs. Job volume varies by time of day and client demand.

Compared to Other GPU Monetization

Method Typical Earnings Notes
Crypto mining (ETH-era)Ended 2022Proof-of-stake killed GPU mining
Crypto mining (alt-coins)$0.10–0.30/hrVolatile, power-hungry, depreciates hardware
Generic GPU rental$0.20–0.50/hrCommodity pricing, race to bottom
Simulacrum$0.60/hrStable demand, hardened containers, no crypto wear

Supported Hardware

16 GB+ RAM recommended. More RAM allows your node to handle larger projects.

Other Requirements

Payouts

You can track your pending and paid earnings in real-time through the Simulacrum Node GUI or the operator dashboard.

How Earnings Are Calculated

Each job is billed per-second at the flat $0.75/hr rate. When a job completes on your node:

Reputation & Proof of Compute

Every operator starts with a reputation score of 100. The Proof of Compute system periodically verifies that your node is honestly computing frames. Maintaining high reputation ensures you continue receiving jobs.

As long as your node is running normally and not tampered with, you will pass all proof challenges automatically.

FAQ

What happens if a job fails or the client cancels?

You are paid for the compute time actually used. If a client cancels mid-job, you receive 80% of the compute consumed up to the cancellation point. If a job fails due to your node disconnecting, you receive nothing for that job.

Do I need special security setup?

No. Docker security is configured automatically by the Setup Docker button. All containers run with hardened security flags (--cap-drop=ALL, --no-new-privileges, memory and PID limits).