GreenPT Sustainability
Sustainability overview
Building a more sustainable future through measured energy and efficient AI infrastructure.
Every request's energy is measured directly from GPU hardware, not estimated. Every carbon number uses the live carbon intensity of the grid running your workload. No annual averages. No projections. We report the physical carbon intensity of the grid our servers actually run on, not the carbon we could claim by buying renewable energy certificates elsewhere. The atmosphere doesn't distinguish between a certificate and a smokestack. Neither do we.
No offsets. No greenwashing. Here's how.
-
Measured, not claimed.
Every request's energy is measured directly from GPU hardware, not estimated. Every carbon number uses the live carbon intensity of the grid running your workload. No annual averages. No projections.
-
The right model for the job
Not every question needs a heavyweight model. We match each task to the most efficient model that can handle it.
-
Leaner models, same results
Through compression and quantization, we make the AI models themselves smaller and more efficient, so every interaction costs less energy from the start.
-
Real numbers, not certificates.
We report the physical carbon intensity of the grid our servers actually run on, not the carbon we could claim by buying renewable energy certificates elsewhere. The atmosphere doesn't distinguish between a certificate and a smokestack. Neither do we.
Measured, not claimed
Green AI shouldn't stop at a border
We started GreenPT in Europe because that's where the strictest privacy rules are, and where renewable energy infrastructure is furthest along. But sustainable AI shouldn't be a European privilege.
The honest reality for North American infrastructure: not every grid is clean. Some regions run largely on hydro or nuclear. Others still burn a lot of fossil fuel. We won't paper over that with certificates. We measure what your request actually consumed, where it actually ran, and we show you the number, per prompt.
Same platform. Same energy indicator on every chat. Just closer to home, and honest about what that means today.
Model Efficiency
Model Efficiency & Future Vision
Our commitment to efficient AI and sustainable innovation.
Optimized Model Selection
We deliberately use smaller, more efficient models to reduce computational requirements without sacrificing capability.
-
Computational Efficiency
Using lower-parameter models reduces computational requirements for the tasks they can handle well.
-
Specialized Agents
Specialized lightweight agents for specific tasks minimize resource usage.
-
Prompt Optimization
Efficient prompts reduce unnecessary token usage.
Sustainability FAQ
How does your infrastructure reduce environmental impact?
- Measured infrastructure. Energy and carbon measured on every request.
- Efficient model selection. The right model for the job.
- Technical optimisations. Smaller, efficient models.
How can you measure energy per request?
Modern GPUs report their power consumption in real time. For every request, we record the GPU's energy counters when processing starts, track power throughout, and calculate the actual energy consumed. We then apply the current carbon intensity of the grid where that GPU is located, using live data from grid operators. That gives a real per-request carbon number, not a projection or an average. Most AI providers don't do this. If you bill per token, hardware-level measurement isn't in your interest. We think users deserve to know what their AI actually costs.
Why don't you claim 100% renewable energy on the US site?
Because we can't honestly claim it. Not every grid our US infrastructure touches is fully renewable, and buying renewable energy certificates to paper over that doesn't change what actually flows into our servers. Instead, we tell you the truth: the real carbon intensity of the grid your request ran on. Some regions are very low-carbon, others less so. We show you the actual number, per request. That's more useful than a blanket "we're green" claim.
Free for 14 days
Sustainable AI, with the numbers to back it up.
Start measuring your AI's energy use and CO₂ in real time, per request.
Cancel any time.
- GDPR compliant
- CCPA-ready
- No training on your data