NB NuttaButtaInAASoftware & AI Projects
Home/Projects/Orayth

Native AI research

ORAYTH
Build the intelligence.

An independently developed AI research project focused on model training, evaluation, instruction following, runtime intelligence, and the systems required to move from experiments to usable capability.

Research areas

Developed from the system upward.

The public page describes the direction without exposing private training data, internal evidence, checkpoints, or development-only implementation details.

Models

Native training

Independent model lineage, controlled training stages, targeted curricula, and iterative improvement.

Evaluation

Evidence-driven gates

Candidate evaluation and promotion decisions are based on measured behavior rather than version numbers alone.

Runtime

Intelligence systems

Research into how trained capability is exposed through a dependable runtime and application layer.

Research

Instruction & reasoning

Experiments in instruction following, semantic discrimination, context use, identity, and broader capability.

StatusActive research and development
ApproachNative lineage with staged evaluation
Public scopeProject overview, not private research artifacts

Native model research progress · August 31, 2026

Stage 3.5 remains live/default while Stage 4 continues through the FIX9 repair line.

Orayth continues to use its native model lineage. Stage 4 has not been promoted into the live/default runtime. The public status now reflects the completed FIX9 curriculum and later repair/checkpoint work without presenting an unevaluated checkpoint as a promoted model.

Live/default

Stage 3.5 DEV7

Working runtime baseline

Stage 3.5 remains the active/default Orayth model while Stage 4 research continues behind promotion gates.

Last completed strict result

Stage 4 FIX8

Strict score: 0.6875

FIX8 remains the last completed strict Stage 4 evaluation retained for public reporting and did not satisfy promotion requirements.

Curriculum

Stage 4 FIX9

64,000-record true-holdout curriculum built

The grounding-retention and anti-attractor curriculum uses separated train, validation and test template families and remains the current Stage 4 curriculum foundation.

Research checkpoints

FIX9 → FIX12

Repair line continuing

Research-best checkpoints have been observed through later FIX10, FIX11 and FIX12 repair work, but no Stage 4 candidate is being presented as promoted until strict evaluation and promotion gates are satisfied.

Promotion status: Stage 4 remains research-only. Stage 3.5 remains live/default.