>Pricing and Capacity, Stated Honestly

Pricing is metered from the same credit ledger as ParalleliX AI (Part VIII.7); 1 credit = 1 $PRLX. The network runs open-source models (currently 7B-class, for example qwen2.5:7b). The Compute API is a cheap parallel executor for bulk independent sub-tasks, not a frontier model, and it is not represented as matching frontier quality.

Capacity is real and finite: it is the set of nodes online at request time. Large batches queue against that capacity; GET /v1/models and GET /v1/usage report what is actually available rather than promising unbounded scale. Metering is custodial, performed by the coordinator, the same trust model as ParalleliX AI v1.