FSI
Syllable alphabet and phonotactic substrate
Private operating layer
The operating layer for Bantu language intelligence: FSI, tone, concord, equations, cartridges, validation, and model adapters held in one runtime architecture.
Operating thesis
BantuOS treats dictionaries as generated artifacts, not the source of the language. The source is the process: syllables, tone, noun class, concord, morphemes, constraints, and templates that generate valid forms and reject invalid ones.
Runtime architecture
The BantuOS core is deterministic, auditable, and cartridge-driven. Model adapters sit above it for contextual ambiguity resolution, but they do not replace the symbolic engine or override its validation rules.
Ecosystem role
Syllable alphabet and phonotactic substrate
Sound-layer evidence and layered orthography direction
Concord registry and agreement intelligence
Formal template and equation registry
Studio-grade bounded generation domain
Domain cartridge pattern for controlled vocabulary
Governance, IDs, API discipline, release semantics
Build status
Architecture console, pages, JSON APIs, deployment docs.
implemented in this appJSON exports from FSI, UCM, templates, and source manifests.
nextSyllable parser, phonotactic validator, selected phonology, tone slot trace.
plannedClass_ID agreement API aligned with UCM.
plannedFirst template family and generated noun examples.
plannedRejection-record schema and first BC:N rules.
plannedHoch comparison subset and discrepancy report.
plannedAmbiguity schema, JSONL training format, evaluation plan.
plannedProve it free, validate it on your own data, or license the whole program.
Score your model against the foundational layer — the Alphabet Test and the L26 Lite suite, with saved results. Self-serve, no cost.
Three languages you choose — and everything we hold for them. Measured on your own held-out data.
Every product, every language, the full consented corpus — and everything curated while you're subscribed.