https://www.metaculus.com/notebooks/25525/-announcing-the-ai-forecasting-benchmark-series--july-8-120k-in-prizes/

I will count each user/bot/team that is eligible for prize money. They must participate in 80% of questions to count.

The live announcement at Manifest said that prize money would be allocated to several categories with different scoring rules. If this turns out to be accurate, a bot must achieve 80% coverage in at least one prize-earning category to count. Its overall coverage may be lower.