Language models
Open weightsQwen for businesses
Qwen is a family of open language models from Alibaba that supports 119 languages and dialects; the dense Qwen3 models are under Apache 2.0. We use Qwen models via the IONOS AI Model Hub – from low-cost Qwen3.5-9B for simple tasks to special models for embedding and reranking in document search.
- Licence
- Apache 2.0 (Qwen3, dense models)
- Operation
- IONOS AI Model Hub (Germany)
- Official website
- qwenlm.github.io
What we use Qwen for
Simple tasks very cheap
Qwen3.5-9B costs €0.10 input and €0.15 output per million tokens at IONOS.
Sophisticated texts
Qwen3.5-397B-A17B for complex tasks when smaller models are not enough.
Better search results
Qwen3-VL-Reranker-8B reevaluates document search hits.
Programming Help
Qwen3-Coder-Next for internal development tasks.
Why Qwen?
Highly multilingual
According to the provider, 119 languages and dialects – helpful for multilingual customers.
Many sizes
From small to very large models, plus embedding, reranker and code models.
Permissive license
The dense Qwen3 models are running Apache 2.0.
Can be used in Germany
Via the IONOS AI Model Hub from German data centers.
What we take care of with Qwen
- Choosing the right Qwen model for your task
- Connection via the IONOS AI Model Hub
- Use of rerankers and embeddings in document search
- License check per model version
Implementation and consulting can often be subsidised through German BAFA or INQA funding.
What it is not meant for
The license can differ between model versions; we check it for each version used. How well a Qwen model formulates your German texts should be shown by a test with your own examples.
Our solutions with Qwen
Frequently asked questions about Qwen
What is Qwen?
+−
A family of open language models from Alibaba in many sizes, plus special models for code, embeddings and reranking.
Is Qwen open source?
+−
According to the provider, the dense Qwen3 models are published under Apache 2.0. The license should be checked for each specific version.
Can Qwen speak German well?
+−
Qwen is designed for many languages. How well it formulates for your tasks is best shown by a comparison test with your own examples.
How much does Qwen cost at IONOS?
+−
Depending on the model, from €0.10 (input, Qwen3.5-9B) to €3.60 (output, Qwen3.5-397B-A17B) per million tokens, net, as of October 2026.
Works together with
Product names and logos mentioned are trademarks of their respective owners. bettersorted uses this software and operates it for clients; there is no official partnership.
A 30-minute conversation, free of charge.
A 30-minute consultation, free of charge. We’ll take a look at your processes and give you our honest opinion on whether automation is worthwhile.