About TreadLightlyAI
Why
TreadLightlyAI started from a question: what would it look like if an AI tool were transparent with you?
That means showing the energy cost of every conversation. It means giving you real control over how much AI you use. It means putting 10% of all revenue toward AI safety, through the Future of Life Institute, disbursed every six months.
Have questions or feedback? We read every message. Reach us at hello@treadlightly.ai
The Models Behind TreadLightlyAI
Green Tiny, Green Small & Green Docs (Infomaniak)
Two models, under three names, run on Infomaniak's Swiss cloud on certified renewable electricity (certifiers on our AI Transparency Card). Green Tiny and Green Small are the same model, Gemma 4 31B Instruct from Google DeepMind, with different limits. Green Small is the default; it also summarizes long conversations and web search results. Green Tiny allows answers half as long, takes less conversation history, and has no Reasoning option; it writes conversation titles and evaluates prompts. Green Docs (BGE Multilingual Gemma 2 9B) powers document search only.
- What is sent: Your message text and conversation history. Attached images are sent to the model as image data. Text and PDF attachments are sent as extracted text, never as the original file. Conversation text is also sent for title generation and prompt evaluation.
- Why Infomaniak: CPU-optimized inference, certified renewable electricity, and a Geneva data center (D4) built to send all of its waste heat into the district heating network.
- Data handling: Infomaniak is a Swiss company subject to Swiss data protection law. Subject to Infomaniak's terms.
Green Large (GreenPT)
GreenPT is a European AI provider. It states that every request runs in data centers on certified renewable electricity, with the certification held by the data center operators. It has not said which data center serves Green Large. Green Large (GPT-OSS 120B) is available for chat on Starter, Builder and Advanced, handles prompt improvement, and backs up title generation when Infomaniak is unavailable. GreenPT is the only provider we use that reports how much energy a request took, so it is the reference point our estimates for every other model are calibrated against.
- What is sent: Your message text and conversation history when using Green Large. Text and PDF attachments are sent as extracted text, never as the original file. Conversation text for prompt improvement and title generation. Green Large does not accept images; when you attach one, a green model that does accept images handles the response.
- Voice input: If you use the microphone, your audio is streamed to GreenPT to turn it into text. We keep no copy. GreenPT reports no energy figure for this step, so it is not counted.
- Data handling: GreenPT states it does not use your data to train models. Subject to GreenPT's Privacy & Terms.
Haiku & Sonnet (Anthropic)
Claude Haiku 5.5 and Claude Sonnet 5.5 are available for chat on the Advanced plan. Anthropic ranks first of nine (C+) on the FLI AI Safety Index, Summer 2026.
- What is sent: Your message text, images as image data, text and PDF attachments as extracted text, the active mode's system prompt, and the current session's conversation history.
- What is not sent: Your email, account details, or data from other conversations.
- Training: Anthropic's commercial terms bar it from training models on what we send.
- Electricity: Anthropic does not disclose its energy source.
Environmental Methodology
How we get the numbers
GreenPT returns the energy each request used in its API response, in watt-milliseconds, on streaming and non-streaming calls alike. Green Large is the one model in our lineup that runs there, so it is the one model whose figures are provider-reported. Everything else is estimated from token counts.
Our default model used to run on GreenPT, which measures. Since August 2026 it runs on Infomaniak, which does not, so most of what you see is now an estimate rather than a measurement. Estimated values start with a ~.
Read the full energy methodology for the formula, the per-token rate behind every model and why it is set where it is, the two measurements those rates are calibrated against, what we deliberately do not estimate, and how uncertain the whole thing is.
Waste heat is still energy
Infomaniak's Geneva D4 data center is built to send its waste heat into Geneva's district heating. We still count the electricity it draws in full: reusing the heat does not reduce it.
Our approach
We do not claim to be carbon neutral or zero emission.
For a one-glance summary, see the AI Transparency Card.
Data Practices
| Topic | Details |
|---|---|
| What we store | Conversations, usage metrics, and environmental data: measured where GreenPT reports it, estimated for every other model. All data is tied to your account and deletable by you. |
| AI training | We do not use your conversations to train AI models. |
| Conversation deletion | You can delete conversations at any time. Hard deletion within 30 days. Backups may persist up to 90 days. |
| Analytics | We use PostHog for product analytics, only with your consent. PostHog may collect page views and clicks. Session recording is turned off: we never capture session replays. Subject to PostHog's Privacy Policy. |
| Infomaniak processing | Chat on Green Tiny and Green Small, plus title generation, prompt evaluation and summarization, use Gemma 4 31B Instruct; document search uses Green Docs. Both run on Infomaniak's Swiss cloud. Infomaniak is subject to Swiss data protection law (nFADP). |
| Third-party providers | Your queries are processed by Anthropic, GreenPT, and Infomaniak. Plain web searches and search summaries send your query to Brave Search, as does automatic research on the green models. On Haiku and Sonnet, Anthropic runs the search. Cloudflare stores the files you upload and our database backups. Their respective privacy policies apply to that processing. See our Privacy Policy for full details. |
How This Compares
ChatGPT, Claude.ai, Gemini and Perplexity show you no energy figure for your own conversations. How we compare
A data center's yearly renewable claim says nothing about how much energy your conversation used. We show an estimate of that, conversation by conversation.
We offer Claude for work the green models cannot match. Anthropic ranks first of nine (C+) on the FLI AI Safety Index(opens in new tab), Summer 2026, and publishes its safety evaluations(opens in new tab).
The open-weight models we offer, which run on certified renewable electricity, come from Google DeepMind (C) and OpenAI (C). Both clear the bar we publish on the AI Transparency Card. Neither is a safety leader.