Sustainable AI Group Launched Energy Tracking Dashboard
The new tool provides comparative energy usage data for proprietary AI models to help organizations monitor consumption.
Updated on Sept. 30, 2026 in Artificial Intelligence
Live Poll
Would you choose an AI model based on its energy efficiency rating?
The Sustainable AI Group has released a dashboard and the CLEER dataset to estimate the energy consumption of closed AI models. The project aims to provide transparency in an industry where providers typically do not disclose detailed system power requirements.
Why it matters
Detailed energy data remains hidden from the public, making it difficult for organizations to understand the environmental impact of their software choices. This dashboard provides the necessary metrics for businesses to make informed decisions when routing workloads to specific models.
Kimi 3 consumes 88 Watt hours per 1-2 hour agentic session, while Claude Fable 5 uses 76.2 Watt hours, a value equivalent to 183 seconds of air-fryer operation. Furthermore, agentic sessions utilize 27 times more energy than standard chat sessions.
The players
Sustainable AI Group
This research organization focuses on measuring and mitigating the environmental impact of artificial intelligence technologies.
Etsy
The online marketplace collaborated with the research group to help refine the methodology used for tracking energy consumption.
The details
The research suggests that larger models consume nearly four times more energy than their lighter counterparts, with some cheaper models unexpectedly showing higher consumption than expensive alternatives. The methodology involved comparing proprietary systems to open models of similar capability that had undergone experimental testing.
Timeline
The Sustainable AI Group launched its research firm in early 2026.
The energy dashboard and dataset were released on September 30, 2026.
The Tech Race
This effort represents a shift toward mandatory transparency in the AI sector, mirroring early movements in cloud computing to standardize energy reporting. It positions model efficiency as a core competitive metric, effectively forcing companies to address the carbon footprint of their underlying infrastructure.
Organizations and developers can now utilize these metrics to make more sustainable routing decisions for their AI workloads. This helps users reduce their overall carbon footprint by selecting models that align with efficiency standards rather than just raw performance capability.
The takeaway
Transparency in AI infrastructure is becoming a critical component of corporate sustainability efforts. Readers should monitor these metrics to ensure that their software choices align with environmental impact goals.
What happens next
The Sustainable AI Group plans to incorporate water consumption metrics into the dashboard in a future update.
Further reading
For more context on how industry standards are evolving, visit our Artificial Intelligence section.
Source note: This article includes information reported by Computing.
Live Poll
Would you choose an AI model based on its energy efficiency rating?
