ai-sustainability

claude-carbon: track the carbon footprint of your Claude Code sessions

CLI tool that automatically tracks environmental impact of AI training runs with a simple dashboard.

⚡ Compare AI models by Accuracy × Cost × Carbon — RTX 5090 benchmarks reveal 4-bit quantization wastes energy on small models

An AI-powered construction waste analyzer that calculates CO2 savings and circular economy scores using Google Gemini.

tracking AI system failures with solar weather

PyTorch benchmark harness comparing full / fine-tuned / quantised NLP models on accuracy, latency, memory, and energy per 1,000 predictions. Produces accuracy-vs-emissions trade-off curves for stakeholder consumption.

EcoTwin Intelligence is an AI-powered Digital Carbon Twin platform that analyzes lifestyle activities, estimates carbon emissions, generates sustainability scores, forecasts future impact, and provides actionable recommendations for greener living.

Configurable Python tool for estimating and comparing CO2 emissions of AI workloads across AWS, Azure, GCP, and on-premise scenarios — using regional grid carbon intensity and hardware power profiles. Built on CodeCarbon.

Prototype 3D globe of AI resource burdens (React Three Fiber)

Data center and ai sustainability

Carbon aware LLM routing framework balancing task complexity, grid carbon intensity, latency, accuracy, and carbon budgets.

A manifesto defining AI as intelligence bound within the biosphere.