Understand Generative AI Environmental Footprint
Discover our impact assessment methodologies and open-source tools for developers.
Discover EcoLogits
What we do
We are dedicated to understanding and mitigating the environmental impacts of generative AI through rigorous research, innovative tools, and community engagement.
- Advanced Methodologies inspired by Life-Cycle Assessment (LCA).
- Open-Source Tools free for personal and commercial use.
- Raising awareness on GenAI environmental impacts.
- Helping developers and organizations arround the globe.
Why It Matters
The rapid evolution of generative AI is reshaping numerous industries and aspects of our daily lives. While these advancements offer some benefits, they also pose substantial environmental challenges that cannot be overlooked. At GenAI Impact, we address these critical issues head-on for several key reasons:
- Environmental Responsibility: GenAI models consumes considerable energy, during training and inference, contributing to GHG emissions.
- Resource Efficiency: Beyond energy, generative AI systems impacts resource usage including water and materials.
- Future Viability: Addressing indirect impacts and rebound effects of global generative AI deployment.
Discover EcoLogits
EcoLogits tracks the energy consumption and environmental footprint of using generative AI models through APIs. Some of the key features are:
- Environmental impacts and energy consumption of LLMs at inference.
- Works with major providers: OpenAI, Anthropic, Mistral AI and more.
- Open-Source licence for personal and commercial use.
EcoLogits Calculator
EcoLogits Calculator is pedagogical web app to estimate the environmental impacts of example LLM prompts for many different models and providers.
- Choose your provider, model, prompt and get the environmental impacts.
- Learn about energy or GHG emissions equivalents of your LLM usage.
- Deep dive into the modeling and better fit to your use case in the expert mode.
- Based on the open-source EcoLogits python library.