As of Thursday evening UTC, the sixth-dimensional primitive in Prometheus7's substrate-paradigm architecture remains in validation. The institute's internal publication record shows that as of late May 2026, a training run designated v4 — the third attempt at validating the router-over-callables primitive — was active. Whether that run completed, and what it produced, has not been reported publicly.
What has been documented is the roadmap it sits inside. A paper ingested to the institute's corpus on 25 May 2026 lays out a twelve-rung dimensional ladder, each rung a new compositional primitive, each primitive adding one kind of operation the substrate algebra did not previously admit. The fifth-dimensional primitive — the substrate routing manifold underlying the Tree of Life model lineage — was recorded as validated on 16 May 2026. The sixth was in its first validating run two days later.
The paper is concrete about what each primitive is. The 6D primitive routes hidden state to one of K small neural sub-modules called callables. The trunk specializes; the callables specialize separately; the router decides which callable contributes at each timestep. The architectural claim is that subsequent generations can absorb new 6D primitives without retraining prior ones, through what the paper calls the bound-axis mechanism.
From there the ladder extends forward in time and in abstraction. The 7D primitive, targeted for the second week after the 6D launch, routes over sets of callables rather than individual ones — parallel compositional reasoning rather than sequential specialist selection. The 8D primitive routes across grammars: not which specialist, not which coalition, but which vocabulary to operate within. The 9D primitive routes across worlds, making the substrate multi-substrate-aware. Each is accompanied by a stated falsification condition — a specific mode in which the new primitive could collapse back into the one below it.
The tenth primitive is what the paper calls the resolution point. The universal-unbinder. The architecture's claim at 10D is that the substrate becomes a universal object: something holding all specifics in superposition, unpacking any of them through relation. The paper maps this to the universal Turing machine in computability, to the holographic principle in physics, to the category of all categories in category theory. The target date given in May was August-September 2026 — meaning, by the dateline of this report, that window has arrived or is arriving.
Whether the 10D primitive has been attempted is not recorded in available material. The corpus carries the roadmap; it does not carry a completion report. What can be said accurately is this: the institute published a schedule in May, that schedule named this month as the target for its most architecturally significant milestone, and the 6D validation that was supposed to precede it was still in progress when the paper was written.
The 11D and 12D primitives are described as research dimensions — work for a small community over years, not for one training run. They are noted but not scheduled.
The ladder is the claim. What has been witnessed is the fifth rung, and the attempt at the sixth. Everything above that is a published intention.