Compounder

One slider. Two lines on a chart. The crossover point. Drag “tokens per day” up and watch the cloud cost line and the local cost line extend across your horizon. Where they cross is when local pays back — concrete, sharable, one screen.

The math is the same as /cost-vs-cloud; the difference is the time dimension. You see not just what you'd pay this month but when the cumulative gap matters.

Configure the scenario

Pick a cloud provider, a local rig, and the volume slider. URL updates as you change — share by copying.

§ Tokens per day500,000 tok/day

10K

20M

§ Cumulative cost over time

cloud burns linearly · local front-loads + amortizes

Cloud / day: $0.000

Local / day: $0.000

Upfront (est.): $0

Break-even: mo 0

Cloud provider

Horizon

Your local hardware

Local model (Q4_K_M assumed)

Price figures are our planning estimates (launch MSRP, or a used-market estimate for older cards), not live prices. Check the current price before you buy.

Output token share (rest is input)30% output · 70% input

Cloud (per day)

—

Local (per day)

pick rig

Upfront (est.)

hardware only

Break-even

immediate

local beats cloud

§ Verdict at 500,000 tokens/day

Pick a local hardware + model to see the crossover analysis.

Where to go from here

Cost vs Cloud →

Per-prompt cost math. Paste an actual sample, see the provider matrix instead of the time-axis chart.

TCO calculator →

Tune assumptions: utilization, electricity rate, cloud equivalent rate, amortization horizon — the inputs feeding this chart.

Stack Builder →

Crossover looks favorable? Build the actual rig — GPU + runtime + models + install script.

CC Decoder →

Want the volume number to be real? Paste a Claude Code transcript; the decoder tells you the actual tokens/day you're running.