AI carbon footprint in EU North - Finland: emissions per token and per query
Factor version soma-ai-ef-2026.10-dataset-v3.0.0. Last updated . Location-based grid averages, not renewable contracts. Estimates from published benchmarks.
Part of the SOMA AI Emission Factors cluster.
How much CO₂ does an AI query emit when served from EU North - Finland?
A 1210-token query to a class B model (gpt-4o) served from EU North - Finland emits about 0.014 g CO₂e, and a million tokens emit 0.0102 kg CO₂e.
Rank 3 of 16 for class B, lowest carbon first (full table). 1.4x the level of EU North - Sweden, Asia Pacific - Singapore is 3.9x higher.
Which grid supplies the electricity in EU North - Finland?
EU North - Finland maps to Finland, a national grid, at 81.3 g CO₂e per kWh for 2023. That is 0.3 times the EU-27 average (236 g) and 0.2 times the global average (483.2 g). The source is Ember yearly electricity data, CO2 intensity 2023, 81.33 g/kWh (lifecycle, IPCC AR5).
Cloud regions on this grid, 1 in all (1 Google Cloud).
| Provider | Region id | Region name | Location | Mapping confidence | Mapping note |
|---|---|---|---|---|---|
| Google Cloud | europe-north1 | europe-north1 | Hamina, Finland | high |
Which grids have an intensity close to Finland?
| Difference g | Grid | Code | g CO₂e per kWh | Ratio |
|---|---|---|---|---|
| 5.9 | New Zealand | NZL | 87.2 | 1.073 |
| -6.9 | Tajikistan | TJK | 74.4 | 0.915 |
| 10.6 | Venezuela (Bolivarian Republic of) | VEN | 91.9 | 1.130 |
| 15.0 | Brazil | BRA | 96.3 | 1.185 |
| 15.6 | Kenya | KEN | 96.9 | 1.192 |
| -22.4 | Uganda | UGA | 58.9 | 0.724 |
| -26.7 | Malawi | MWI | 54.6 | 0.672 |
| -28.0 | France | FRA | 53.3 | 0.656 |
| 30.4 | Austria | AUT | 111.7 | 1.374 |
| 31.8 | Uruguay | URY | 113.1 | 1.391 |
Emission factors for EU North - Finland by model class
| Model class | Electricity kg | Hardware kg | Construction kg | Training kg | Total kg CO₂e | g per query | Water L | mL per query |
|---|---|---|---|---|---|---|---|---|
| Class A (small) | 0.00177 | 0.00027 | 0.00014 | 0.00060 | 0.003 | 0.0037 | 0.099 | 0.14 |
| Class B (mid-size) | 0.00580 | 0.00101 | 0.00046 | 0.00293 | 0.010 | 0.014 | 0.324 | 0.45 |
| Class C (frontier) | 0.02159 | 0.00388 | 0.00169 | 0.01284 | 0.040 | 0.053 | 1.208 | 1.7 |
| Model class | Central | Class range low | Class range high | High case | B200 | Development |
|---|---|---|---|---|---|---|
| Class A | 0.002782 | 0.001889 | 0.00383 | 0.004055 | 0.002523 | 0.002414 |
| Class B | 0.0102 | 0.007358 | 0.01208 | 0.0144 | 0.008101 | 0.007906 |
| Class C | 0.04001 | 0.02212 | 0.05437 | 0.05569 | 0.04075 | 0.02944 |
How much does siting matter for a model served from EU North - Finland?
Electricity is 56.9% of the class B total here (class A 63.6%, class C 54%), with hardware at 9.9%, construction 4.5% and training 29%. A cleaner grid shrinks the electricity term, while the other 43.1% stays put.
The example company (2,323 million class B tokens a year) emits 24 kg CO₂e from EU North - Finland. Moving to EU West - France would save 4.6 kg (20%) or EU North - Sweden would save 7.1 kg (30%).
| Region | Grid g per kWh | Class A | Class B | Class C |
|---|---|---|---|---|
| EU West - France | 53.3 | 1.42 | 4.65 | 17.3 |
| EU North - Sweden | 38.4 | 2.17 | 7.11 | 26.5 |
How much water does AI usage consume in EU North - Finland?
A class B query of 1210 tokens from EU North - Finland consumes about 0.45 mL of water and a million tokens 0.324 litres, or 0.123 litres without the water attributed to hydropower.
| Input | Value |
|---|---|
| On-site WUE, L per kWh | 0.01 |
| WUE denominator | IT load (server energy) |
| WUE basis as published | consumption |
| Withdrawal to consumption | 0.6667 |
| EWIF, L per kWh | 4.542 |
| EWIF without hydro, L per kWh | 1.712 |
| AWARE factor | 2.17 |
| Confidence | 2023-site |
| Source | Li et al. 2025 (Microsoft); Finland* (projected, site under construction as of July 2023); as of July 2023 |
| Model class | On site | Grid electricity | Total | Total without hydro | AWARE-weighted on site |
|---|---|---|---|---|---|
| Class A | 0.000181 | 0.0989 | 0.0991 | 0.0374 | 0.000394 |
| Class B | 0.000594 | 0.324 | 0.324 | 0.123 | 0.00129 |
| Class C | 0.00221 | 1.21 | 1.21 | 0.457 | 0.0048 |
Where do these numbers come from and how do I cite them?
Figures per query are one plain exchange, before the Tier 2b reasoning uplift. Electricity term. GPU energy per token from the ML.ENERGY Leaderboard v3, a facility overhead of 1.20, and grid carbon intensity from EPA eGRID 2023 and Ember 2023. Embodied hardware, data-centre construction and amortised training terms are documented on the methodology page. Water combines on-site cooling water (Li et al. 2025) with the water embedded in grid electricity (Reig et al. 2020). The modelled range of the models inside a class, around the class value, is −41% to +48% for class A, −39% to +24% for class B and −67% to +55% for class C.
Dataset: AI Inference Emission and Resource Factors for Corporate GHG Inventories (SOMA AI Emission Factors (SAEF)), Llopis-Garcia, G., SOMA AI, 19 October 2026, CC BY 4.0, concept DOI 10.5281/zenodo.20443585, which resolves to the latest version (this version 10.5281/zenodo.23371447, earlier version 2 values at 10.5281/zenodo.22767475). Method: Accounting for Purchased AI Inference in Corporate GHG Inventories: A Token-Based Method, arXiv:2606.10660, under peer review.
Partners can fetch any row from the AI factors API, and the whole table is a CSV download.
Cite as "0.0102 kg CO₂e per million tokens, class B, EU North - Finland, SAEF, 19 October 2026" and link somaai.earth/ai-footprint/factors/eu-finland.