5. Comparing CAPEX and OPEX

Historical scenario; not a current rate card. This comparison uses old assumptions such as $0.005/node-hour, modeled node performance and (in some chapters) fixed burns or token-price projections. The native public mainnet is not yet active. These are not verified current payouts, price forecasts or like-for-like benchmarks; see Pricing Methodology and Status.

CAPEX Comparison

Metric	Centralized xAI Cluster	GNUS.ai Decentralized Network
Hardware Investment	Approximately $6 billion for a 100k–chip cluster	Minimal nodes are owned by end users
Data Center Infrastructure	Requires extensive investment (cooling, racks, etc.)	Software–based; no centralized data center needed
Overall CAPEX	Extremely high	Negligible (only software and coordination CAPEX, e.g., $2M over 8 years)

OPEX Comparison

Metric	Centralized xAI Cluster	GNUS.ai Decentralized Network
Compute Payment (Hourly)	$1.75 per chip × 100,000 chips ≈ $175,000 per hour	$0.005 per node × ~6.7 million nodes ≈ $33,500 per hour
Annual Compute Cost	~$1.53 billion per year	~$293 million per year (paid directly to node operators)
Electricity & Cooling	High estimated ~$61 million per year for electricity alone	Borne by individual node operators (minimal impact on GNUS)
Network & Storage Costs	Additional expenses (~$100 million/year)	Minimal—local connectivity and distributed storage
Total Annual OPEX	Approximately $1.7 billion per year	On the order of $2–$10 million per year (true OPEX for coordination)

Note: For GNUS.ai, the $293 million figure represents the total “compute payment” distributed in tokens, which is a pass–through payment to node operators and not an overhead cost for the GNUS network itself.