STACKIT AI Model Serving

Use leading GenAI models easily and securely

AI chip and hardware acceleration for artificial intelligence and machine learning

Use leading GenAI models easily and securely

STACKIT AI Model Serving gives you easy access to the latest GenAI models in a secure environment on the data-sovereign STACKIT Cloud. This component of our Data & AI Platform enables you to use various Large Language Models (LLMs) with maximum data sovereignty. Your data and queries are neither stored nor used to train models. You choose the LLM that suits you and get a seamless user experience when integrating it into your applications through our API.

Register for a free trial access

Test our STACKIT AI Model Serving free of charge and get easy and secure access to leading LLMs - thanks to STACKIT Cloud with maximum data sovereignty.

All registrations are manually checked by us. If you qualify, you will receive your personal token to use STACKIT AI Model Serving within 24 hours. The trial period ends automatically after 30 days.

Register now

Applications of STACKIT AI Model Serving

The following use cases, among others, can be realized with STACKIT AI Model Serving:

Create, edit and analyze texts efficiently

With an LLM, you can create and edit high-quality texts in a matter of seconds. It is also possible to summarize existing texts. By integrating RAG (Retrieval Augmented Generation), you can even analyze your own documents or external data sources. You can find a corresponding template here.

Take your service to the next level with an intelligent chatbot

Take your service to the next level with an intelligent chatbot
Connect your own chatbot to an LLM of your choice and supplement the data using our open-source STACKIT RAG template. This allows you to enter relevant content from your internal knowledge database or subject-specific databases as parameters for queries to the LLM. This allows you to improve the quality of your automated customer service while retaining control over all the data entered.

Optimize processes

Automate recurring tasks and increase your process efficiency. Counteract the shortage of skilled workers and relieve your workforce by automating routine tasks. For example, automated suggestions can be made for article descriptions for online stores.

STACKIT AI Model Serving functions

Integration of the desired LLMs into your own applications via API
High compatibility with existing systems for seamless integration thanks to corresponding interfaces
Extension of the LLM's database with company-specific data possible using the open-source STACKIT RAG template

To the STACKIT AI Model Serving v1.4 effective November 6, 2026

To the STACKIT AI Model Serving service certificate

Advantages of STACKIT AI Model Serving

Operate GenAI applications cost-effectively with our token-based billing.
Stay vendor-independent and flexible by using open source models and open interfaces.
Access innovative LLMs without sacrificing data security and sovereignty.
Simple integration of our STACKIT AI Model Serving into your existing environment enables you to implement it quickly.
Benefit from easy access and an intuitive application - from token creation to the exchange of models.

STACKIT AI Model Serving price table

Model	Type	Region	Input token price (per 1M token)	Output token price (per 1M token)	SKU
Google Gemma 4 31b	Chat	EU01	0.45€	0.65€	Input: Model Serving-llm-standard-plus-EU01 Output: Model Serving-llm-standard-plus-EU01
GPT oss 120b	Chat	EU01	0.45€	0.65€	Input: Model Serving-llm-standard-plus-EU01 Output: Model Serving-llm-standard-plus-EU01
GPT oss 20b	Chat	EU01	0.15€	0.25€	Input: Model Serving-llm-standard-input-EU01 Output: Model Serving-llm-standard-output-EU01
Meta Llama 3.3 70B Instruct FP8	Chat	EU01	0.45€	0.65€	Input: Model Serving-llm-standard-plus-EU01 Output: Model Serving-llm-standard-plus-EU01
Qwen3 VL 235B A22B Instruct FP8	Chat	EU01	1.50€	1.75€	Input: Model Serving-llm-standard-premium-EU01 Output: Model Serving-llm-standard-premium-EU01
Qwen3.6 27B	Chat	EU01	0.45€	0.65€	Input: Model Serving-llm-standard-plus-EU01 Output: Model Serving-llm-standard-plus-EU01
Qwen3.8 27B	Chat	EU01	0.45€	0.65€	Input: Model Serving-llm-standard-plus-EU01 Output: Model Serving-llm-standard-plus-EU01
e5 mistral 7b	Embedding	EU01	0.02€	-	Input: Model Serving-embedding-standard-input-EU0
Qwen3 VL Embedding 8B	Embedding	EU01	0.08€	-	Input: Model Serving-embedding-plus-input-EU0

Man with cell phone in hand and laptop open.

Contact form

Would you like to find out more about STACKIT cloud solutions? Do you have questions about cloud services, would you like an individual price estimate or would you like to find out more about our company? Use our contact form to get in touch with our experts directly. We will be happy to help you!

To the contact form