Ren-Research/Making-AI-Less-Thirsty
[Preprint] Making AI Less ''Thirsty'': Uncovering and Addressing the Secret Water Footprint of AI
This project helps data center operators and AI researchers understand and quantify the often-overlooked water consumption associated with training large AI models. By inputting information about AI training jobs, it estimates their hourly carbon efficiency and on-site water usage effectiveness. This tool is designed for sustainability officers, environmental impact analysts, and researchers in AI or data center operations.
No commits in the last 6 months.
Use this if you need to calculate the environmental footprint, specifically water and carbon usage, for training large AI models and want to identify more sustainable training schedules or locations.
Not ideal if you are looking to optimize the water usage of small, local machine learning tasks or general computing, rather than large-scale AI model training in data centers.
Stars
31
Forks
3
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Apr 07, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Ren-Research/Making-AI-Less-Thirsty"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
NVIDIA/earth2studio
Open-source deep-learning framework for exploring, building and deploying AI weather/climate workflows.
mllam/neural-lam
Research Software for Neural Weather Prediction for Limited Area Modeling
chengtan9907/OpenSTL
OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning
NVIDIA/earth2mip
Earth-2 Model Intercomparison Project (MIP) is a python framework that enables climate...
microsoft/aurora
Implementation of the Aurora model for Earth system forecasting