M0
Truth and contracts: Reconciled legacy truth, canonical metadata, immutable gates, and a controlled whole-model baseline.
M1
Protocol and native substrate: Versioned gRPC protocol, exact identities/artifacts, pinned upstream, reproducible builds, ShardEngine, and fake worker.
M2
Dense vertical proof: Dense ranged execution, parity, local state, worker integration, and GGUF inputs to existing routing.
M3
DeepSeek V4 Flash alpha: Pinned V4 adapter around upstream llama.cpp, real route certification, and pre-locked alpha decision with MTP off.
M4
Performance and beta hardening: Batching, backpressure, recovery, scale certification, optimization, MTP, and hardware matrix.
M5
Release and maintenance: Reproducible packages, upstream collaboration, beta decision, and sustainable recertification.