Why does Kaspa's total supply look slightly different across explorers and dashboards?

Small differences in quoted Kaspa supply figures come from the way each node independently tracks the halving threshold, which can introduce a tiny amount of noise into total emission. Kaspa's halving timing is based on a DAA (Difficulty Adjustment Algorithm) score — a block count that each node calculates on its own. Because Kaspa is a blockDAG, multiple blocks can be produced simultaneously at different parts of the network, so two nodes can briefly disagree on whether the halving threshold has been crossed. During those transition windows (up to 18 seconds), some miners earn the old reward while others earn the reduced one. According to the source, this noise would hardly exceed tens of thousands of KAS across the entire supply — a rounding difference rather than a meaningful discrepancy.