Prices

Every model, ≤ 50% of Venice list

Repriced at :00 / :15 / :30 / :45 UTC from live Venice-backend market cost: cost + five points of Venice list, hard-capped at half of list. Type to filter, click a column to sort.

updated
Model Apex in Apex out Cache in Venice in % of Venice
loading…

usd per 1M tokens. “% of Venice” = our input price ÷ Venice list input. Free-tier models are excluded from this board.