30+ models. One API. Zero compliance headaches.

The LLM Serving Service gives your teams OpenAI-compatible access to the leading AI models: open-source models hosted by Telekom in Germany, frontier models from OpenAI, Anthropic, Google, and Mistral under their own published terms. One contract, one endpoint, one straight answer for your auditor.

Request a trial key, or try preview models free on any plan.

The same problems reappear in every customer conversation.

It's rarely the models that slow you down. It's everything around them.

Ten pilots. Ten stacks. Ten different issues.

Each team builds AI on its own: separate vendors, separate systems, time wasted on multiple compliance checks. Effort put in, but your organization is still miles away from scaling. Nothing is reusable, and every new use case pays the full setup cost again.

AI adoption is stuck in legal & compliance review.

Every AI provider triggers the same questions: Is our data safe? What happens to our inputs? What if a security incident occurs? Months pass, competitors ship.

The best model changes every quarter. Your vendor list can't.

Your teams want GPT today, Claude tomorrow, an open-source model for the sensitive workload. Onboarding each provider separately means a new procurement round every time.

We built LLM Serving to solve them.

Time to shift gears: from pilot to production.

Change the base URL. Keep everything else.

LLM Serving speaks the OpenAI API standard. If your code works with OpenAI, it works with us: same SDKs, same request format, one configuration change. Behind that one endpoint sits the full catalogue. Switch models by changing one string — including switching between a US frontier model and a Germany-hosted open-source model, per request, per use case.

from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AIFS_API_KEY"],
    base_url="https://llm-server.llmhub.t-systems.net/v2",
)

response = client.chat.completions.create(
    model="gpt-oss-120b",
    messages=[{"role": "user", "content": "Hello from sovereign Europe."}],
)

Copy code

Pick the right model for each task.

Choose your data-handling regime: restrict your selection to Telekom-operated models in Germany, or use the full catalogue.

Hosted by Telekom in Germany (T-Cloud)

open-source models including GLM 5.3 Flash and Qwen 3.8 27B — both now in production — alongside GLM 5.2, GPT-OSS 120B, Gemma 4, NVIDIA Nemotron 3 and Mistral Small, plus German-language embeddings and Whisper speech-to-text. Your prompts are not persistently stored, not visible to Telekom or third parties, and never used for training. A prefix cache holds recent requests briefly to speed responses up — you can switch it off.

Frontier models from third-party providers

GPT-5 family on Azure, Claude on Google Cloud or Azure, Mistral Medium in the EU, Gemini 3 on Google Cloud infrastructure worldwide. Operated by the respective provider under its own terms, contracted through Telekom.

One catalogue. 30+ models.

Chat, reasoning, coding, vision, speech-to-text, image generation and, in preview, typed agent decisions. New models are added continuously.

Built for regulated work

Public-sector and enterprise teams build on AIFS and T-Cloud today.

EU AI Act.
Delivered in line with Telekom's obligations under the EU AI Act.
Sovereign and secure infrastructure.
AIFS runs on T-Cloud public: BSI C5 Type 2 attested and ISO 27001 certified Telekom infrastructure.
GDPR-compliant.
With data processing agreements available and a German contractual counterparty.
Transparent by design.
Server locations, terms of use, and licensing provisions are published per model.

Experiment, iterate, scale.

One platform from your first API call to organisation-wide rollout.

LLM Serving

Production-ready service. 99.5% availability during supported business hours, backed by service credits.
Self-service portal. Generate and manage API keys, track token usage by key and model, add users — no ticket required.
Scalable by design. Ready to match strict technical requirements, with tenant-separated, dedicated instances where needed.

LLM Serving

Need more?

Our team provides professional services around the product: dedicated model instances for exclusive capacity and additional models on request. Fine-tuning and deployment in your own environment are also available as professional services where a use case justifies them.

Use cases

Common starting points. Yours may be different.

Sovereign coding assistants

Point OpenCode, Cline, or your internal tooling at Germany-hosted coding models. AI pair programming without sending your source code outside the EU.

Customer service at scale

Run your voicebots and chatbots on one governed endpoint with per-model compliance choices and usage tracking per API key.

Document processing pipelines

Extraction, classification, and summarization over sensitive documents, with vision-capable models and the option to keep processing on Telekom infrastructure in Germany.

Speech and transcription

Whisper hosted in Germany: transcribe calls, meetings, and recordings containing personal data without a US data transfer discussion.

Something else in mind?