Market price is what you pay on Gondola, quoted from the cheapest offer listed right now, including the platform fee of up to 2 percent. It moves as lenders join, reprice, or drain, and is capped near Venice's own list price: at worst about what Venice charges direct. Venice list and official provider prices are those vendors' posted prices for a direct customer, before any fee of their own. Text comparisons assume a typical message of 300 input plus 400 output tokens. Image and video comparisons are tier-matched: both sides price the same resolution, quality, and clip settings (hover a percent for the exact basis); where tiers cannot be matched, the provider price is shown without a percent. Models with a Cached price bill repeated prompt prefixes at that rate; caching is automatic except on models whose Cached tip says they need cache_control markers. Latency is the median response time for a short fixed request we send hourly, so the numbers are comparable across models.