Savings calculator — what the measured reductions do to your bill

What that is worth on your own bill

Put your own figures in and the measured reductions get applied to them — every line names the benchmark behind it, and nothing here is a number we invented about you. If you would rather see the engine run before you price it, the benchmark is on the front page and takes about ten seconds.

YOUR NUMBERS

Put In Your Figures. See What The Measured Ratios Do To Them.

We do not know what you spend, so we do not guess. You supply the baseline; we supply the reductions we have measured, and every line below names the benchmark it came from.

READING ENGINE…

Composed cost reduction84.54%0 under-served

Composed compute saving54.5%0 violations

Guards780 of 780across 190 guards

ComputedPublishedengine unreachable

Your industry

Typical corporate facility overhead means every watt of compute removed takes roughly half a watt of cooling and distribution with it.

Picking an industry only changes the starting numbers below and the assumed facility overhead. The measured reductions are the same for every industry on this list — they are a property of the engine, not of who is running it.

Your baseline

Routable inference spend

$/ yr

Annual spend on inference that could be routed. Not your whole cloud bill.

Servers

Servers in your own facility running this workload.

Watts per server

Your average draw per server. Ours is not a substitute for yours.

Power cost

$/ kWh

Your contracted electricity price.

Cooling & facility (PUE)

Facility overhead (PUE): total site draw ÷ compute draw. 1.0 shows compute alone. Yours, not ours.

Include the performance modules25% of saving plus capacity gained — the lower rate, on the wider base.

Inference spend avoided$10.14M / yr84.54% composed cost reduction, measured on the published composed cost benchmark.

Compute energy saved10,742 MWh / yr54.5% composed compute saving, measured on the published composed compute benchmark with zero violations. Compute draw only — not facility power.

Energy cost avoided$1.50M / yrCompute energy saved at your own $0.14/kWh.

TOTAL SITE ENERGY AVOIDED16,113 MWh / yrCompute energy saved × your PUE of 1.50: removing compute removes the cooling and distribution that served it. ESTIMATED — the multiplier is your facility's, the compute reduction is measured.

TOTAL POWER BILL AVOIDED$2.26M / yr16,113 MWh at your $0/kWh. Includes the cooling and distribution avoided, not just the compute.

Total annual saving$11.65M / yrInference spend avoided plus compute energy cost avoided. Both are money not spent — they share a unit and may be added.

Capacity gained, valued$14.37M / yr54.5% compute saved ⇒ 2.20× PER UNIT OF MODELLED COMPUTE, valued at your own inference rate. Not tokens per second and not a hardware measurement: the saving is measured in relative compute units against a declared model, and this is that same measurement read the other way — which is why it is shown separately and added only once.

Combined value$26.02M / yrMoney not spent plus work now doable without buying hardware. Added once, as value — never summed into a single percentage.

WUBBERY share (25%)$6.51M / yr25% of the combined value above — the rate falls from 35% because the performance modules widen what it is charged on. If nothing is saved and no capacity is freed, nothing is owed.

Net to you$19.52M / yrCombined value less the 25% share.

The number we could have shown you instead

$14.00M / yr

Adding the levers instead of composing them on the remainder gives 116.64% — more than the spend itself. We publish the gap rather than the bigger number.

What this is and is not

These are estimates from your inputs, not a measurement of your systems.
The ratios are measured on our benchmark traffic. Your mix is not our mix — the shadow replay runs against your own logs and returns your number, without changing what you serve.
Compute energy is compute draw only. Cooling, PUE and your AI share of IT load are inputs only you have.
Cost avoided and capacity gained have different denominators, so they are never combined into a single percentage. They are added once, as money, only to price the share — and both halves stay on screen so the addition can be checked.
Capacity is a modelled figure: the compute saving is measured in relative compute units against a declared model, not in tokens per second on hardware.

Want your real number rather than an estimate? The shadow replay runs against your own logs and changes nothing you serve.

Check the ratios yourself: GET /api/wubbery/proof — no key, no input, computed on request.

The same reductions across 32 industries — and the percentage barely moves

The same engine, 32 different bills

Every row below uses the identical measured reduction — 84.54% off cost, 54.5% less compute. Nothing here is an industry-specific claim, because there is no such thing: the saving is a property of the engine, not of who is running it. What changes down the table is the size of the bill it is applied to, and how much cooling your building adds on top.

Industry	Modules	PUE	Cost avoided	Capacity value	Site MWh	Power $	Combined	You keep	We earn
Hyperscale / cloud providerData centres & infrastructure	82	1.15	$33.82M	$47.91M	76,865	$6.15M	$81.73M	$61.30M	$20.43M
AI platform / inference providerData centres & infrastructure	64	1.25	$16.91M	$23.96M	21,484	$2.15M	$40.86M	$30.65M	$10.22M
Enterprise data centreData centres & infrastructure	78	1.50	$10.14M	$14.37M	16,113	$2.26M	$24.52M	$18.39M	$6.13M
Pharmaceutical / life sciencesRegulated	56	1.60	$7.61M	$10.78M	6,951	$904k	$18.39M	$13.79M	$4.60M
Financial servicesRegulated	29	1.55	$6.76M	$9.58M	4,995	$649k	$16.35M	$12.26M	$4.09M
Telecom / ISP / internetData centres & infrastructure	33	1.45	$5.92M	$8.38M	5,538	$609k	$14.30M	$10.73M	$3.58M
Colocation / managed hostingData centres & infrastructure	36	1.45	$5.07M	$7.19M	4,154	$665k	$12.26M	$9.19M	$3.06M
Automotive / mobilityIndustry & operations	23	1.55	$4.65M	$6.59M	4,884	$586k	$11.24M	$8.43M	$2.81M
Insurance / actuarialRegulated	25	1.55	$3.38M	$4.79M	2,331	$303k	$8.17M	$6.13M	$2.04M
Aerospace / space & defenceIndustry & operations	25	1.70	$2.96M	$4.19M	2,435	$292k	$7.15M	$5.36M	$1.79M
Government / public sectorRegulated	18	1.65	$2.54M	$3.59M	1,654	$248k	$6.13M	$4.60M	$1.53M
Retail / e-commerceIndustry & operations	22	1.55	$2.54M	$3.59M	2,072	$290k	$6.13M	$4.60M	$1.53M
Research / scientific computingIndustry & operations	50	1.50	$2.37M	$3.35M	4,010	$441k	$5.72M	$4.29M	$1.43M
Energy / utilitiesIndustry & operations	27	1.60	$2.11M	$2.99M	2,062	$144k	$5.11M	$3.83M	$1.28M
Security / cyber operationsRegulated	22	1.60	$1.86M	$2.64M	1,375	$192k	$4.50M	$3.37M	$1.12M
Healthcare / clinicalRegulated	50	1.70	$1.69M	$2.40M	812	$122k	$4.09M	$3.06M	$1.02M
Mining / resourcesIndustry & operations	29	1.95	$1.52M	$2.16M	1,396	$279k	$3.68M	$2.76M	$919k
Chemical / process industryIndustry & operations	25	1.85	$1.35M	$1.92M	1,214	$121k	$3.27M	$2.45M	$817k
Games / real-time 3DSoftware & services	39	1.40	$1.27M	$1.80M	1,337	$187k	$3.06M	$2.30M	$766k
Logistics / transportIndustry & operations	21	1.60	$1.01M	$1.44M	962	$125k	$2.45M	$1.84M	$613k
Media / creative productionSoftware & services	27	1.50	$845k	$1.20M	859	$137k	$2.04M	$1.53M	$511k
Manufacturing / industrialIndustry & operations	18	1.85	$761k	$1.08M	530	$64k	$1.84M	$1.38M	$460k
Legal / professional servicesRegulated	37	1.80	$507k	$719k	180	$31k	$1.23M	$919k	$306k
Sport / events & venuesSoftware & services	19	1.60	$507k	$719k	344	$52k	$1.23M	$919k	$306k
Construction / built environmentIndustry & operations	16	1.95	$423k	$599k	261	$42k	$1.02M	$766k	$255k
Real estate / propertySoftware & services	38	1.60	$380k	$539k	94	$14k	$919k	$690k	$230k
SaaS / software productSoftware & services	45	1.40	$338k	$479k	94	$14k	$817k	$613k	$204k
Agriculture / food productionIndustry & operations	16	1.90	$338k	$479k	190	$25k	$817k	$613k	$204k
Education / schools & universitiesSoftware & services	46	1.75	$254k	$359k	117	$18k	$613k	$460k	$153k
Hospitality / food serviceIndustry & operations	19	1.85	$211k	$299k	53	$9k	$511k	$383k	$128k
Non-profit / NGOSoftware & services	33	1.80	$76k	$108k	13	$2k	$184k	$138k	$46k
Small business / startupSoftware & services	29	1.90	$34k	$48k	5	$980	$82k	$61k	$20k
One of each (32)							$290.40M	$217.80M	$72.60M

Two things worth reading off this table. The spread from the smallest row to the largest is roughly a thousandfold — which is why a single calculator with a single default was never going to speak to everyone. And facility overhead moves opposite to size: manufacturing at 1.85 and legal at 1.80 get proportionally more back per watt removed than a hyperscaler at 1.15, because they are paying more to cool every watt they run.

These baselines are illustrative, and they are not your numbers. They are a plausible shape for each kind of operation, chosen so the calculator opens on something recognisable rather than on a hyperscaler’s fleet. The percentages are measured; the dollars are arithmetic applied to a bill we guessed. Put your own figures into the calculator above and the same measured ratios produce your number instead of ours.

Why our rate falls as the deal widens, shown rather than claimed. Those same 32 customers on 35% of the saving alone would pay us $42.05M. On 25% of saving plus capacity they pay $72.60M — and every one of them is better off, because the second deal only charges for capacity that was actually handed back. A lower percentage of a bigger, real number. If we find nothing, both columns are zero.

84.54% COMPOSED COST REDUCTION · 54.5% COMPOSED COMPUTE SAVING (2.2× CAPACITY) · ENERGY IS ESTIMATED, NOT METERED · VERIFY: GET /v1/public/proof

THIS SHOWS COST AVOIDED. THE OTHER HALF — CAPACITY GAINED, THE SAME MEASUREMENT READ AS THROUGHPUT — IS IN THE CONSOLE, ALONGSIDE PER-INDUSTRY BASELINES AND YOUR OWN LIVE RATIOS. Create an account to use it. Free, and the shadow run that produces your real numbers changes nothing you serve.