susty: sustainability advice that knows what it costs

Got questions on what's more sustainable?

Ask about diet, travel, home energy, or anything else.

How we counted this

Estimates, not promises. Token counts come from the API; everything after is modelled.

This conversation

Answered by	claude-sonnet-4-6
Extended thinking	off
Exchanges	0
Tokens sent (context)	0
Tokens written back	0
Chip energy, sent @ 0.05 mWh/token	0 Wh
Chip energy, written @ 0.5 mWh/token	0 Wh
Cooling & overhead ×1.12	0 Wh
Vercel hosting & network per exchange	0 Wh
Your device	0 Wh
Electricity drawn	0 Wh
Grid intensity	473 g/kWh
Total	0.00 g CO₂e

Grid carbon intensity

Answer with

Reading this on

What to do about it

Offsetting this costs under a hundredth of a cent.

At roughly $100 per tonne for verified carbon removal, this conversation is a rounding error. These three actions each cover it thousands of times over:

Wash clothes in cold water, around 200 g saved per load.
Thermostat down 2°F, roughly 300 kg a year in a typical heated home.
One beef meal a week swapped, on the order of 200 kg a year.

To put it in perspective

The footprint of asking is small next to the footprint of what you do with the answer.

Notes

We do our best to make these calculations as accurate as possible. Some of it has to be estimated, so here's where the numbers come from and where we're guessing. The full method has all of it.

One thing we don't know precisely, but would be great to find out: which specific servers are answering your questions, in a data center somewhere else, running on their own electricity. We've assumed they're close to you and on the same energy sources, which is an estimate. We'd love to know exactly which servers are doing the work and what that data center draws, but that isn't public information right now. Our goal is to run this locally on a sustainable server ourselves.

The sum itself is simple: energy used × how dirty the electricity is. Extended thinking, where it is on, is counted inside the tokens written back: it is charged at the same rate, and it is the thing that decides whether an exchange costs once or several times over. The total also includes the device you're reading on, not just its screen, which is how the published methods count it. But only the part you were actually here for: typing and scrolling are the evidence, and once neither has happened for a while we stop counting. It's an estimate, and the row shows it next to how long the page has simply been open. And longer chats cost more per answer, because we send the whole conversation back each time.

Methodology

Energy per token (0.05 / 0.5 mWh) Luccioni et al. 2023: Power Hungry Processing, Watts Driving the Cost of AI Deployment
Data centre overhead PUE ×1.12 Uptime Institute 2023 Global Data Center Survey: global average power usage effectiveness
Great Britain: live, by region NESO Carbon Intensity API. Half-hourly actuals for 14 distribution regions, so GB readings here are current rather than averaged. Open and keyless; nothing about you is sent with the request.
United States: hourly by region EIA-930 hourly generation by fuel type, turned into an intensity with EIA's own per-fuel figures (coal 2,257 lb/MWh, gas 976 lb/MWh). Two caveats worth knowing: EIA publishes hourly CO₂ only in bulk spreadsheets, so this is derived from the generation mix rather than read off, and the feed itself runs about a day behind. We therefore take the same hour on the previous day, which tracks the daily curve far better than their newest figure would. That one is the middle of last night. Interregional imports are ignored. Needs a free EIA key; without one we fall back to the annual figures below.
United States: annual, by eGRID subregion EPA eGRID2023 Rev 2 CO₂e output rates, converted from lb/MWh. Subregions run from 110 g/kWh in upstate New York to 680 on Oahu, so the 384 national average describes almost nobody. Used when no live reading is available.
Everywhere else: national annual averages 2024 country figures, cross-checked against Ember's Global Electricity Review 2025. Annual means, so they miss both time-of-day swings and regional spread.
Carbon removal cost ~$100/tonne Frontier 2024 State of Carbon Dioxide Removal: durable CDR market pricing
Dark mode display power (3–9% at normal brightness) Dash & Hu, MobiSys 2021: How much battery does dark mode save? An accurate OLED display power profiler for modern smartphones. Measured across several OLED phones: ~42% of total phone power at 100% brightness, but only 3–9% at the 30–50% brightness in normal use.
Display power (modelled, not measured) The browser can't tell us what panel you have, so the figures above come from the picker and the study beside it, not from your device. The OLED phone delta is calibrated into that 3–9% band; laptop and monitor rows scale it by panel area and are extrapolation, since only phones were measured. LCD is modelled as zero because a backlight's draw is content-independent. Local dimming and mini-LED are a real exception we don't try to quantify. Screen-on time is genuinely measured, and pauses when the tab isn't visible.