Skip to content

GreenPT is now available on LLMTR

The LLMTR logo above the GreenPT logo, joined by a small cross, on a light background.

GreenPT's EU-hosted, open-weight models are now available through LLMTR, a Turkish AI gateway. Same LLMTR API, with inference in the EU.

GreenPT’s European inference is now available through LLMTR, a Turkish AI gateway. Teams that already use LLMTR can now choose EU-hosted, open-weight models through the same API, with requests processed in the EU and never forwarded to the companies that built the models.

What is LLMTR?

LLMTR is a Turkish AI gateway built around one idea: where your data is processed matters. It gives teams one OpenAI-compatible API for hundreds of models from the big providers. Alongside those, it has long offered models hosted in Türkiye itself, for teams that need their data to stay in the country or want models trained for Turkish.

What’s new

LLMTR is adding a European option. Next to its Türkiye-hosted models, teams can now choose open-weight models that run on GreenPT’s infrastructure in the EU. LLMTR keeps the list of available models on its EU Hosted models page.

“We started offering Türkiye-hosted models because data sovereignty matters to many of our users. Our partnership with GreenPT allows us to extend that choice to Europe, giving developers access to EU-hosted AI models through the same API they already use.”

İskender Çevik, founder, LLMTR

Why it matters

With most AI services, a request to a model goes to the company that built it, wherever that company processes its data. Open-weight models work differently: we run them on our own servers, so your prompts stay with us, in the EU. Nothing goes back to the original model maker, and nothing is used to train models.

For teams working with internal documents, RAG systems, codebases or enterprise applications, that makes the processing location a choice instead of a surprise.

How to start

LLMTR is compatible with the OpenAI SDK, so it comes down to a base URL, a key and a model string.

Before your first request, turn on data-transfer consent for GreenPT in your LLMTR account settings.

from openai import OpenAI

client = OpenAI(
    base_url="https://llmtr.com/v1",
    api_key="YOUR_LLMTR_API_KEY"
)

response = client.chat.completions.create(
    model="greenpt/glm-5.3-eu",
    messages=[
        {"role": "user", "content": "Hello from LLMTR!"}
    ]
)

print(response.choices[0].message.content)

Two gateways, one idea

LLMTR is the second AI gateway to add GreenPT this month, after Opper AI. Both started from the same question: where does the data actually go? We think more teams will ask it.

Running a router or gateway yourself? GreenPT is OpenAI-compatible, so adding us is mostly a base URL and a key. Get in touch at mail@greenpt.ai.

“Where your data is processed should be a choice you make, not something you find out later. LLMTR has been giving Turkish teams that choice for a while. We’re happy to add a European option to it.”

Robert Keus, co-founder and CEO, GreenPT

Frequently asked questions

Is my data processed in Türkiye when I use a GreenPT model through LLMTR?

No. The models run on GreenPT's infrastructure in the EU, so inference takes place there.

Which GreenPT models can I use through LLMTR?

LLMTR keeps the current list on its EU Hosted models page.

Do I need to enable anything before my first request?

Yes. Turn on data-transfer consent for GreenPT in your LLMTR account settings. You do this once, before your first request.

I already use the GreenPT API. Do I need to change anything?

No. Your existing integration keeps working as it does today. LLMTR is an extra route to the same models, not a replacement.

Back to all posts