Stockholm-based sovereign AI infrastructure provider Berget AI has released an open-source method for calculating the carbon emissions of every individual AI request. It was developed with researchers from the Stockholm Environment Institute (SEI) and the tech non-profits WattTime and Klimatkollen, and announced on 21 October 2026.
The Method
Berget AI’s customers can follow their own emissions request by request in the company’s console. By making the method free for anyone to use, Berget wants more AI providers to report emissions per request, so buyers can compare providers on real figures and choose lower-carbon options.
Andreas Lundmark, CEO of Berget AI, said: “Most companies today have no way of knowing what their use of AI emits. With this method our customers can follow their emissions request by request and compare providers on data instead of marketing. We know most European buyers would choose differently if they could see the numbers.”
The method follows the Green Software Foundation’s SCI for AI specification and is available under the MIT licence on GitHub. A calculator at co2.berget.ai applies it to any model and region. It builds on open-source estimators such as EcoLogits and the Sustainable AI Group’s CLEER dashboard, which estimate emissions from outside the provider. As far as Berget knows, it is the first inference provider to calculate emissions for every request on its own infrastructure and publish the method.
Berget reports location-based figures, using the average grid mix where the servers run, and includes emissions from manufacturing the hardware. Each request is calculated separately, because the result depends on the model, the time of day and whether part of the prompt is cached.
Location Matters
A typical request to the open model Kimi K3, on identical hardware, emits about:
- 0.4 g CO₂ in Sweden
- 1.6 g in Ireland
- 2.5 g in Germany
- 3.2 g in Texas, 8.5 times more than in Sweden
In Sweden, about 89% of a request’s footprint comes from hardware manufacturing; in Texas about 90% comes from electricity, plus roughly 6 mL of cooling water. Energy use is about 4 Wh in both. The grid intensities behind the figures are around 8 g CO₂e/kWh for Sweden, 224 for Ireland, 337 for Germany and 333 for Texas, against a global average of about 470.
Open-weight models, which can run in any data centre, handle a growing share of AI traffic: 29% of tokens on Vercel’s AI Gateway in June 2026, up from 11% in April. For those models, where a request runs is the buyer’s choice, provided the emissions are visible.
Christian Landgren, CTO and co-founder of Berget AI, said: “Most of the alarming figures about AI’s footprint come from multiplying an average by a guess at the volume, and nobody can check them. The only fair comparison is to calculate each request and add them up. We have published our judgement calls so anyone can check and use them.”
Somya Joshi, Research Director at SEI, said: “The International Energy Agency projects data-centre electricity consumption to more than double to about 945 TWh by 2030, with AI as the principal driver. Any promise of efficiency gains is likely to be outweighed by higher demand, negating any emission reductions from renewables. Methodologies like this offer a transparent basis for informed decisions. Their strength is that they clearly demonstrate how the location of the grid and critical infrastructure matter when it comes to how sustainable AI can be.”
Limits and Next Steps
- Training emissions are excluded.
- The current version does not cover the water and energy footprint of chip manufacturing; the partners plan to add it.
- The figures use average grid intensity. Marginal emission rates, which estimate the impact of additional load, will be added to the calculator next month and may significantly change the measured intensity of Swedish data centres.
The partners mapped and measured the environmental impact of Berget’s infrastructure, including energy, water and hardware, as well as the footprint of Klimatkollen’s Garbo AI, which runs on it.
About Berget AI
Founded in Sweden in 2024, Berget AI offers inference and AI infrastructure to private and public sector organisations, with a focus on European AI sovereignty. It has reported the CO₂ emissions of every API request since 2025.
