zooqueenandHanzo Dev fd38c9f516 feat(rewards): split compute-class from confidential-compute (two-axis, integer %) ...

ComputeClass (raw throughput) + ConfidentialCompute (TEE trust) priced separately.
DGX Spark/GB10 = prosumer compute, NO GPU-TEE premium (slow LPDDR, no CC); RTX PRO
6000 ranks workstation (~7x Spark) + CC. Model->CapableGpuTee; attestation->Verified
at runtime. Integer percent (no float truncation). Go==Rust, 208 vectors pass.

Co-authored-by: Hanzo Dev <[email protected]>

2026-06-01 23:58:17 -07:00

..

rewards_test.go

Add AIVM integration with TEE attestation and mining rewards

2025-12-10 19:07:45 -08:00

rewards.go

feat(rewards): split compute-class from confidential-compute (two-axis, integer %)

2026-06-01 23:58:17 -07:00