FAQ
Are there any additional network costs?
No. All data traffic is included in the price.
Which models are available?
Currently, the models OpenAI gpt-oss-20b, Qwen3.8-27b, BAAI bge-m3, and BAAI bge-reranker-v2-m3 are available.
Will additional models become available?
Yes, we are continuously working on making more models available.
How is data processed?
The prompt sent by the user is transmitted encrypted via TLS and processed directly by the model. The generated response is then sent back to the user. At no point is data stored, read, or shared with third parties.
How is billing handled?
Billing is based on the number of processed input and output tokens.
- Input tokens: per 1 million tokens
- Output tokens: per 1 million tokens
How many API keys are included?
You can create and name an unlimited number of API keys.
Can I integrate a RAG (Retrieval-Augmented Generation)?
Yes, this is possible in principle. Feel free to contact us – we will support you in building a suitable solution.
How can I integrate the API?
The API can be integrated via a standardized OpenAI-compatible interface. You can find detailed documentation in the Tutorial section.
Is there a test environment?
Yes, you can first try the model in our free OpenWebUI Chat Interface.
Where are the servers located?
Our systems run in German data centers in accordance with the ISO-27001 standard.
Is there a data processing agreement (DPA)?
Yes, upon request, we will provide a DPA in accordance with the GDPR.
How do I get support in case of problems?
By email at nws@netways.de.
Can I manage access for my organization's members?
Yes, you can configure permissions via the web interface MyNWS.
Does the API support streaming responses?
Yes, streaming is available to receive responses in real time.