Language models
Open weightsgpt-oss for businesses
gpt-oss is a family of open language models from OpenAI under the Apache 2.0 license, consisting of gpt-oss-120b and gpt-oss-20b. We use gpt-oss-120b via the IONOS AI Model Hub from German data centers, especially for more demanding tasks and AI agents with tool calls.
- Licence
- Apache 2.0
- Operation
- IONOS AI Model Hub (Germany)
- Official website
- huggingface.co/openai/gpt-oss-120b
What we use gpt-oss for
AI Agents
Tool calls and structured outputs make gpt-oss suitable for agents in n8n.
More difficult document issues
When medium models don't answer accurately enough.
Reading data
Reliable JSON outputs for automations.
Local testing
gpt-oss-20b runs with 16 GB of memory even on small hardware.
Why gpt-oss?
Apache 2.0
Commercial use, adaptation and distribution without special requirements.
Adjustable amount of thought
Low, medium or high – depending on the difficulty of the task.
Cheap at IONOS
€0.15 input and €0.65 output per million tokens (net, as of October 2026).
No data flow to OpenAI
No data goes to OpenAI via IONOS or operated locally.
What we take care of with gpt-oss
- Test gpt-oss with your tasks
- Connection via the IONOS AI Model Hub or local operation
- Use in agents and workflows
- Reconciliation of thought effort and costs
Implementation and consulting can often be subsidised through German BAFA or INQA funding.
What it is not meant for
gpt-oss is not the model behind ChatGPT; the latest closed OpenAI models can be stronger in some tasks. The model expects its own response format ("harmony") that the operating software must support.
Our solutions with gpt-oss
Frequently asked questions about gpt-oss
What is gpt-oss?
+−
A family of open language models from OpenAI under Apache 2.0: gpt-oss-120b (117 billion parameters, 5.1 billion active) and gpt-oss-20b (21 billion, 3.6 billion active).
Can I use gpt-oss commercially?
+−
Yes, the Apache 2.0 license allows commercial use, adaptation and distribution if the license notices are retained.
Does data go to OpenAI?
+−
No. Via the IONOS AI Model Hub or on your own hardware, the model runs completely outside of OpenAI.
What hardware does gpt-oss need locally?
+−
According to OpenAI, gpt-oss-120b runs on a graphics card with 80 GB of memory, gpt-oss-20b with 16 GB.
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.