Created by Ben O’Leary Co‑Founder & Chief Quantum Officer, Nirmata Holdings

603‑930‑2131 ben@nirmataholdings.com www.nirmataholdings.com

ATOM · V4 release · formula contract IY-v2.1 · source-grounded

Useful intelligence
per joule.

Not one magic number — a vector. Completed tasks per joule, net dollars per joule, watts per active user, latency at p99, and a bounded index that can be vetoed. Every input carries an evidence tier and a live low/base/high interval.

On the phrase. “Intelligence per joule” is an orchestration objective, not an SI unit. Intelligence has no accepted scalar measure, so this app substitutes a declared, auditable proxy: completed tasks on a named evaluation suite, divided by measured joules inside a declared boundary. Read the substitution as a design choice, not a discovery.

Intelligence Yield index and six desirability rings Concentric arcs. Each arc length is one desirability between 0 and 1; the centre shows the bounded index.

Six rings = six desirabilities. Arc length is the desirability, ring thickness is its weight. Exact values are in the synchronized inspector below — nothing important is read off the picture.

Command

Scenarios are labelled by what they actually are. The default is a measured reference with every delivery-convergence assumption switched off.

Bounded index

Intelligence Yield index
/ 100

IY = 100·exp(Σ wⱼ·ln dⱼ) — a weighted geometric mean in log space, so a near-zero factor cannot be bought back by a strong one. Weights sum to exactly 1. The index is dimensionless and always relative to the frozen reference contract named above; it is never a physical rate.

Physical headline rates

MEASURED reference

Yenergy and Yvalue are per completed task, not per attempt — attempts per completed task (na) is an explicit input, so retries cannot be hidden. Yvalue is allowed to go negative; loss-making configurations are shown as losses.

V4 release · formula contract IY-v2.1 · REF-2026.1

The methodology kernel

One objective. Three mathematical layers.

Every layer below is printed in full, substituted with the values currently loaded, and labelled with what it is allowed to be used for. Only layer C is the authoritative score.

On the two version numbers. V4 is the product release you are looking at. IY-v2.1 is the mathematical formula contract it runs on, frozen under reference REF-2026.1. V4 added this kernel, the baseline observatory and the reference contract; it did not change the index formula, the weights, the frozen targets or the reference basket, so every V3 index value still holds.

    Symbol legend — every term, its unit and its role
    SymbolMeaningUnitCurrentRole / source

    Baseline observatory

    The sourced numbers this model is calibrated against. Every row carries an evidence tier, its workload and boundary, an as-of date and a live link. Full detail, including the cross-source normalisation limits, is in Evidence & Method.

    Per active user and whole-system vector

    Every figure names its unit and its allocation basis. “Per active user” means divided by the declared active-user count — an accounting allocation, not a measurement of one person's hardware.

    Token throughput efficiency Φtoken = Ru / Wu [tokens / J]. Ru = Tu/86400 is the output-token rate per active user [tok/s], and Wu = Eu/24 is allocated average power [W], because watt-hours per day divided by 24 hours is watts. Dividing a rate by a power gives (tok/s)/(J/s) = tokens per joule, and the words “per active user” cancel exactly — which is why Φtoken is an intrinsic property of the serving path while W / active user stays the genuine per-user capacity number. The uncorrected form Tu/Wu is a token count over a power and is never shown as a rate in this app.

    Executive controls

    five dials — a scenario in seconds

    Baseline energy, usage, task success, cache hit rate and serving gain move almost all of the vector. Everything else is available below, and in full in the Measurement Lab.

    Advanced controls latency p99 · throughput @ concurrency · tokens · split prices · privacy · routing · network · attempts/task · rebound

    Delivery convergence

    all default to 0
    These are modelled assumptions about an edge/routing/caching layer, not measurements. They stay at zero until you move them or load a modelled scenario. Nothing in the reference case depends on them.
    Residual energy factor1.000
    residual = x·S_hit + (1−x)(1−P_route)(1−M_serve)  ·  E_adj = (1−f)·E_gross·residual + f·E_gross·(1−N_net)

    Fleet totals

    constant-demand accounting

    Index
    tasks / MJ