Model catalogue · OpenAI

GPT-OSS

Version via EU inference of open models through the Gosign chain: GPT-OSS 120B , model openai/gpt-oss-120b (as of 29 September 2026)

EUR 0.25 per million tokens as of 12 October 2026

Vendor: gpt-oss-120b, as of 29 September 2026

OpenAI's open-weights model - not to be confused with ChatGPT. The answering style carries a related signature, but it can be run on your own premises instead of being called at the vendor; the feature set stays behind the large ChatGPT class. For professional secrets four routes are viable at once: running it on your own hardware means no third party participates, so the disclosure question under section 203 of the German Criminal Code does not arise; in dedicated operation on reserved hardware and with the open models we host ourselves the signed confidentiality undertaking of the respective operator closes the chain. The Open Telekom Cloud also offers the model, under its own standard agreement with its customers. At IONOS the product-specific confirmation of that clause is so far missing.

Available through us

Everything section 203 requires is settled, including the provider's confidentiality undertaking, and nothing is stored at the provider, apart from a technical cache of a few minutes for Claude (zero retention). You do not have to clarify anything further.

The operator bills separately - input EUR 0.15 · output EUR 0.55. The figure above is the blend of the two, weighted three quarters input to one quarter output, as it arises in a firm. Charged at the vendor's list price, without a mark-up.

The providers set these prices; they can change at any time, and our figures may contain errors. The list price in force at the time is authoritative.

What it is good for

  • Open weights without licence restrictions - operation in Europe or on your own premises
  • Familiar answering style for teams used to OpenAI models
  • Available on several section 203-capable routes, including a German provider

Where it stops

  • Performance below the current proprietary OpenAI models
  • The state of the weights is ageing - no successor has been announced
  • References have to be verified; no legal fine-tuning

Status through us

The same model, procured differently, is a different legal position. The status above refers to procurement through us; processing takes place in Germany.

Which route carries it, with which instrument and under which condition, we set out in full to the firm before it commissions us - section 203(4) requires that the chain can be checked, not that it is published. The full matrix, with the sources behind every route, can be requested in the German catalogue after confirming your e-mail address, and we go through it in the conversation.

Is GPT-OSS the right fit?

We go through your tasks and the route that carries them - and say plainly where it does not.

Arrange a conversation