Green AI 🌱
A curated overview of resources for reducing the environmental footprint of AI development and usage.
Contributions and pull requests are welcome!
The following tools are designed to calculate the footprint based on information about the choice of algorithms, configuration and hardware.
- d2m [Website] [Source code] – a machine learning pipeline for ML model development with automatic monitoring and tracking of the carbon footprint
Papers
Particularly important papers are highlighted.
- Energy and Policy Considerations for Deep Learning in NLP (Strubell et al. 2019) [Paper]
- Quantifying the Carbon Emissions of Machine Learning (Lacoste et al. 2019) [Paper]
- Green AI (Schwartz et al. 2020) [Paper] [Notes]
- Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models (Anthony et al. 2020) [Paper]
- Carbon Emissions and Large Neural Network Training (Patterson, et al. 2021) [Paper]
- Chasing Carbon: The Elusive Environmental Footprint of Computing (Gupta et al. 2020) [Paper]
- Green Algorithms: Quantifying the Carbon Footprint of Computation (Lannelongue et al. 2021) [Paper]
- A Pratical Guide to Quantifying Carbon Emissions for Machine Learning researchers and practitioners (Ligozat et al. 2021) [Paper]
- A framework for energy and carbon footprint analysis of distributed and federated edge learning (Savazzi et al. 2021) [Paper] [Notes]
- Aligning artificial intelligence with climate change mitigation (Kaack et al. 2021) [Paper]
- New universal sustainability metrics to assess edge intelligence (Lenherr et al. 2021) [Paper]
- Unraveling the Hidden Environmental Impacts of AI Solutions for Environment Life Cycle Assessment of AI Solutions (Ligozat et al. 2022) [Paper]
- Measuring the Carbon Intensity of AI in Cloud Instances (Dodge et al. 2022) [Paper]
- Estimating the Carbon Footprint of BLOOM a 176B Parameter Language Model (Luccioni et al. 2022) [Paper]
- Bridging Fairness and Environmental Sustainability in Natural Language Processing (Hessenthaler et al. 2022) [Paper]
- Eco2AI: carbon emissions tracking of machine learning models as the first step towards sustainable AI (Budennyy et al. 2022) [Paper]
- Environmental assessment of projects involving AI methods (Lefèvre et al. 2022) [Paper]
- Sustainable AI: Environmental Implications, Challenges and Opportunities (Wu et al. 2022) [Paper]
- A first look into the carbon footprint of federated learning (Qiu et al. 2022) [Paper] [Notes]
- The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink (Patterson et al. 2022) [Paper]
- Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learning (Henderson et al. 2022) [Paper]
- Towards Sustainable Artificial Intelligence: An Overview of Environmental Protection Uses and Issues (Pachot et al. 2022) [Paper]
- Measuring the Environmental Impacts of Artificial Intelligence Compute and Applications (OECD 2022) [Paper]
- Method and evaluations of the effective gain of artificial intelligence models for reducing CO2 emissions (Delanoë et al. 2023) [Paper]
- Making AI Less “Thirsty”: Uncovering and Addressing the Secret Water Footprint of AI Models (Li et al. 2023) [Paper]
- Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training (You et al. 2023) [Paper]
- Chasing Low-Carbon Electricity for Practical and Sustainable DNN Training (Yang et al. 2023) [Paper]
- LLMCarbon: Modeling the End-To-End Carbon Footprint of Large Language Models (Faiz et al. 2023) [Paper]
- Power Hungry Processing: Watts Driving the Cost of AI Deployment? (Luccioni et al. 2023) [Paper]
- A Synthesis of Green Architectural Tactics for ML-Enabled Systems (Järvenpää et al. 2023) [Paper]
- Toward Sustainable HPC: Carbon Footprint Estimation and Environmental Implications of HPC Systems (Li et al. 2023) [Paper]
- Exploring the Carbon Footprint of Hugging Face’s ML Models: A Repository Mining Study (Castaño et al. 2023) [Paper]
- Estimating the environmental impact of Generative-AI services using an LCA-based methodology (Berthelot et al. 2023) [Paper]
- From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference (Samsi et al. 2023) [Paper]
- Power Hungry Processing: Watts Driving the Cost of AI Deployment? (Luccioni et al. 2023) [Paper]
- Perseus: Reducing Energy Bloat in Large Model Training (Chung et al. 2024) [Paper]
- Timeshifting strategies for carbon-efficient long-running large language model training (Jagannadharao et al. 2024) [Paper]
- Engineering Carbon Emissions Aware Machine Learning Pipelines (Husom et al. 2024) [Paper]
- Measuring and Improving the Energy Efficiency of Large Language Models Inference (Argerich et al. 2024) [Paper] [GitHub]
- Green AI: Exploring Carbon Footprints, Mitigation Strategies, and Trade Offs in Large Language Model Training (Liu et al. 2024) [Paper]
- A simplified machine learning product carbon footprint evaluation tool (Lang et al.) [Paper]
- Beyond Efficiency: Scaling AI Sustainably (Wu et al. 2024) [Paper]
- Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems (Miao et al. 2024) [Paper]
- Towards Greener LLMs: Bringing Energy-Efficiency to the Forefront of LLM Inference (Stojkovic et al. 2024) [Paper]
- The Price of Prompting: Profiling Energy Use in Large Language Models Inference (Husom et al. 2024) [Paper]
- Hybrid Heterogeneous Clusters Can Lower the Energy Consumption of LLM Inference Workloads (Wilkins et al. 2024) [Paper]
- Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systems (Wilkins et al. 2024) [Paper]
- AI, Climate, and Regulation: From Data Centers to the AI Act (Erbert et al. 2024 [Paper
- LLMCO2: Advancing Accurate Carbon Footprint Prediction for LLM Inferences (Fu et al. 2024) [Paper
Survey papers
- Evaluating the carbon footprint of NLP methods: a survey and analysis of existing tools (Bannour et al. 2021) [Paper]
- A Survey on Green Deep Learning (Xu et al. 2021) [Paper] [Notes]
- A Systematic Review of Green AI (Verdecchia et al. 2023) [Paper]
- Counting Carbon: A Survey of Factors Influencing the Emissions of Machine Learning (Luccioni et al. 2023) [Paper]
Leaderboards
Organizations, projects and foundations
- Green Software Foundation – non-profit foundation promoting software development with sustainability as a core priority [Website]
- ENFIELD: European Lighthouse to Manifest Trustworthy and Green AI – project for creating a European Centre of Excellence with Green AI as one of the pillars [Website]
Other resources