Show HN: Sunk Cost – How long until a local LLM rig pays for itself?

Sunk Cost, a tool presented on Show HN, helps determine how long it takes for a local LLM rig to pay for itself. Users can adjust assumptions like Electricity, $/kWh, and API speed, tok/s. A key assumption is that API prices will continue to fall. The tool estimates local speed based on memory bandwidth ÷ bytes read per token × when no direct measurements are available, and this estimation is clearly labeled. API speed primarily impacts the time comparison aspect of the analysis.