GreenPT

What is GreenPT?

GreenPT is a Dutch AI company built around two promises: your data is never used to train a model, and the environmental cost of AI is made visible instead of being left off the bill. Its API serves open-weight models, from its own GreenL and GreenR to GLM, DeepSeek, Kimi and MiniMax, through an OpenAI-compatible interface. GreenPT tells its side of the partnership in its own announcement: GreenPT is now available on Eden AI.

What sets it apart is that sustainability is measured rather than claimed. GreenPT reports the energy and carbon of every request, and it builds tools to reduce both, starting with models that simply say less.

GreenPT at a glance

A green alternative, measured rather than claimed

Energy and carbon on every request

Most AI providers publish a sustainability page. GreenPT puts the numbers in the response. According to GreenPT, the energy of each request is read from the GPU hardware that served it, and the carbon figure uses the live intensity of the grid that GPU is connected to, not an annual average or a renewable certificate.

That has a practical consequence: a team that has to report on the footprint of its AI features can log it request by request, next to latency and cost, instead of estimating it once a year.

Through Eden AI, these figures come back in every GreenPT response as an impact object, with inference time, energy in watt-milliseconds and estimated emissions in micrograms of CO2e. No other provider on Eden AI returns this today.

Fewer tokens, less energy

The cleanest token is the one that is never generated. GreenPT's compression models are variants of GLM 5.2 that answer with fewer output tokens at the same price per token: caveman for prose, ponytail for code and honey for both, each in a lite and an ultra intensity.

On its own test set, GreenPT measured output reductions of roughly a third to three quarters against the base model, with most of the saving coming from shorter reasoning and explanation rather than shorter code. Fewer tokens means less compute per answer, and a smaller bill at the same time.

The right model for the job

Not every question needs a heavyweight model. GreenPT's own GreenL and GreenR carry a built-in sustainability system prompt, while their -raw twins carry none, so your own prompt is the only one sent. For new projects, GreenPT points to GLM 5.3 and GLM 5.3 Flash as its recommended chat models.

When should you choose GreenPT?

Choose GreenPT if:

You need to report the footprint of your AI usage. Per-request energy and carbon, measured on the hardware, is the raw material a sustainability report actually needs.

Your answers are long. Coding agents, documentation and support replies are where compression models save the most tokens, and the most energy.

Privacy is part of the requirement. GreenPT does not train on your data and does not keep it beyond short-lived caching, and it only serves open-weight models you can audit.

Consider another provider if:

Your workload is short and high-volume. For classification, routing or extraction, a compression model's built-in instruction costs more input tokens than it saves on output, so use the base model or another provider.

You need a frontier proprietary model. GreenPT serves open-weight models only.

GreenPT pros and cons

GreenPT models on Eden AI

GreenPT's own models

Open-weight models served by GreenPT

This is a selection. GreenPT serves more models through Eden AI, and the live list with current pricing is in the Eden AI catalogue: GET https://api.edenai.run/v3/models is public and needs no key.

What can you build with GreenPT?

AI features with a footprint you can report

Log GreenPT's per-request energy and carbon next to your usual metrics, and turn a vague sustainability goal into a number per feature, per customer or per month.

Leaner coding agents

Point an agent at glm-5.2-ponytail or glm-5.2-honey and it writes less boilerplate and shorter explanations, while keeping code, commands and security warnings intact.

Transcribe, then summarise

Transcribe pre-recorded audio with green-s-pro, then summarise or classify it with GLM 5.3 Flash, all behind the same Eden AI key.

Why use GreenPT through Eden AI?

A green option is only useful if you can adopt it one workload at a time. Through Eden AI, GreenPT is one string in the request, next to every other provider, so you can move your longest-answer workloads to a compression model and compare cost and output without rebuilding anything.

curl https://api.edenai.run/v3/chat/completions \
  -H "Authorization: Bearer $EDENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "greenpt/glm-5.2-honey",
    "messages": [{"role": "user", "content": "Add retry with backoff to this fetch call."}]
  }'

One key, one bill, and switching between GreenPT and another provider for the same model is an edit, not an integration.

Frequently asked questions about GreenPT on Eden AI