NoxHash Energy Profile — Our Power Source, Cost, and Carbon Footprint

@shared_image('about', 'power-substation', 'Industrial power substation feeding the NoxHash data center — the physical interface between Kazakhstan's grid and our fleet')

Where your NoxHash rental power actually comes from

When you rent a miner on NoxHash, the electrons powering it come from an industrial contract on the Kazakhstan national grid. That grid's generation mix — like most grids in Eurasia — leans heavily on fossil fuels. Roughly two-thirds of electricity in Kazakhstan is generated from coal, with natural gas making up most of the remainder, and hydro plus small renewables rounding out the bottom. Our contract draws from this general mix; we do not cherry-pick specific plants or hours.

We do not operate our own renewable generation. We do not purchase renewable energy certificates (RECs) to re-label fossil kilowatt-hours as "green." We do not claim carbon-neutral or net-zero status. Several cloud mining operators do make those claims — often using instruments like unbundled RECs that critics argue do not meaningfully shift the grid. We have chosen not to use that playbook.

What we do deliver is operational efficiency and transparency: a fleet running at roughly half the joules-per-terahash of the global mining average, natural winter cooling that cuts auxiliary HVAC load, a power contract cheap enough to stay viable through bear markets, and an energy profile we are willing to publish rather than obscure. The sections below lay out the numbers.

Kazakhstan grid composition — what's in the mix

Kazakhstan's generation mix has been heavily coal-dominated for decades, a legacy of abundant domestic reserves in the north and east of the country. Natural gas contributes roughly a fifth, used both in combined-cycle plants and combined heat-and-power stations in large cities. Hydro — mostly from large Soviet-era plants on the Irtysh and Syr Darya rivers — provides a meaningful but limited share, and wind and solar are growing off a low base under a national renewables modernization target.

Shares shift year to year — a dry year trims hydro, new wind capacity comes online, some coal units retire — but the shape is stable: fossil-heavy, with a slowly greening edge. Our industrial contract cannot opt out of that composition. When you see the Cambridge Centre for Alternative Finance's country-level carbon intensity data for Kazakhstan, you are looking at roughly the same mix that powers your rental.

Power cost — $0.035/kWh industrial contract

Our industrial electricity contract prices in at approximately $0.03-0.04 per kWh. That range is roughly 3-5x cheaper than residential electricity in the US or EU, competitive with Paraguay's hydro-heavy south and on par with industrial rates in Texas before summer peak pricing kicks in. Low power cost is not itself an environmental claim — it is an operational one. It means the fleet can run through bear markets without forcing shutdowns that would leave miner owners paying for idle hashrate.

Notice the regions that beat or match our price — Paraguay, Quebec, Norway — all draw on hydro, which carries a dramatically lower carbon footprint. We are cheap, but we are not clean. A miner hosted in Quebec at $0.05/kWh is meaningfully greener than one hosted with us at $0.035/kWh. We are not going to pretend otherwise. What we will argue is that fleet efficiency — covered next — is the lever we directly control and the one that determines how many kilowatt-hours flow into each BTC we mine.

Fleet energy efficiency — the lever we actually control

The grid mix is set by national policy; we are a grid customer, not a generator. The fleet, however, is entirely our decision. We choose which ASICs to buy, when to upgrade, and what the average joules-per-terahash of the operation looks like. A more efficient fleet means fewer kWh burned per BTC mined, regardless of where those kWh came from.

The fleet average sits around 15 J/TH. Plenty of operators — especially smaller home or semi-professional setups — still run older-generation hardware at 25-30 J/TH, and a non-trivial number of S9-era units remain in service at ~95 J/TH. Relative to those, our fleet consumes roughly half the electricity per TH/s of hashrate. Put differently: for every BTC mined on our fleet, a comparable S19 Pro-era fleet would burn about 2x the kWh, and an S9-era fleet would burn about 6-7x. Efficiency is the single biggest lever inside an operator's control, and it is where we invest.

@shared_image('about', 'kazakhstan-steppe-wide', 'Open Kazakhstan steppe — the site characteristics that give us cold winters, dry air, and low land cost')

Natural cooling — the one genuine green-adjacent win

Kazakhstan's continental climate delivers winters of -15 degC to -30 degC across most of the country for roughly four to five months of the year, and cool shoulder seasons on either side. For a data center, that is a gift. Outside ambient air carries most of the thermal load through straightforward exhaust-fan designs; mechanical chillers stay off for long stretches. Industry benchmarks put auxiliary cooling load at roughly 20-30% lower than equivalent facilities in warm, humid regions (Texas summer, Paraguay year-round).

This is not a renewable-energy story — it is a thermodynamics story. A kWh you never had to spend on a chiller is a kWh you do not burn at a coal plant. In a coal-heavy grid, cooling efficiency is arguably the single most impactful green-adjacent lever available, because every saved cooling kWh directly reduces fossil generation. It is also the honest answer to the question "why operate in a coal region instead of moving to hydro?" — the cooling saving is real, immediate, and persists across every operating hour.

An honest carbon footprint estimate

Let us do the math openly. Kazakhstan's grid carbon intensity averages somewhere in the range of ~550-600 grams of CO2 per kWh — a figure consistent with a coal-heavy generation mix, although hourly variation is significant. Our fleet average of ~15 J/TH translates to roughly 15 watts per TH/s of continuous draw.

Running one TH/s for a month: 15 W x 24 hours x 30 days / 1,000 = ~10.8 kWh/month per TH/s. At 600 g CO2/kWh, that is about ~6.5 kg CO2 per TH/s per month. Scaling up to our rental tiers:

These numbers are illustrative estimates, not a certified inventory. Actual grid carbon intensity varies hour-to-hour based on which plants are dispatching, and our own plant efficiency varies with inlet temperature and load. We are showing the math rather than hiding it — if a third-party audit later produced numbers 20% higher or 30% lower, we would not be shocked. What matters is the order of magnitude: cloud mining at this fleet efficiency on this grid is a non-trivial but comprehensible footprint, comparable to other energy-intensive personal activities rather than to industrial manufacturing.

What NoxHash is NOT doing — the honest list

To close the loop on transparency, here is what we are explicitly not doing, that some other operators do. We are listing these because the absence is a choice, and you should be able to evaluate that choice:

Some of these are paths we may revisit as Kazakhstan's grid greens up (the national renewables target, if met on schedule, would shift the mix meaningfully by the 2030s) or as our own facility economics change. For now, our position is: efficiency plus transparency, not offsets plus marketing. Reasonable people can disagree with that trade-off; we would rather let you disagree with the actual facts than with a brochure.

Frequently asked questions