I was scrolling through some of the older listings on the Model Hub and noticed something that's easy to miss if you're only looking at the network's totals. A handful of models sitting at the same call count they had in their first week. Nothing about that looks broken. The model loaded fine. Someone bothered to set a price and leave the listing live, so the first call presumably worked. The first call wasn't the part that made me stop. The price was already there. The listing was already live. Someone had clearly expected more than one inference when they set it up that way. A first call proves the model works. It says nothing about whether it's worth coming back to. Those are different questions, and Model Hub's whole earning design depends on the second one, not the first. A builder publishes once and gets paid automatically every time the model gets used again. That mechanism does nothing for a model that only ever gets used once. The dashboard does not tell me which shape I'm looking at.
How many builders ever earn from their model a second time without having to go find users themselves?
$UB On that day, the price was still going up! I looked at the data: the contract positions volume increased by more than 10 million within 24 hours, and both the quantity and the price rose together—there is still more upward momentum!
$RE The fees for the Asan Three package and the size of holdings both fell sharply at the same time, and they cannot rise above them; it will inevitably drop to a lower level. Welcome to your joining the Air Force!