We measure regularly against the same test set. If a value drops below the criterion, we fix it.
We test a successor model with your test set first and only switch once it passes.
Monitoring, maintenance and support sit with us. Your IT does not need a new hire for it.
A share of the project price, billed monthly. Code, data and documentation belong to you.
Providers switch off older model versions on fixed dates, and the successor behaves differently. If nobody retests, you find out through complaints.
Vendor information: OpenAI deprecation schedule for older modelsEven without a version change, a hosted model can shift noticeably within a few months. Researchers at Stanford and Berkeley measured this on the same tasks.
Chen, Zaharia, Zou: How is ChatGPT's behavior changing over time?, Stanford/UC Berkeley 202343%
A supplier changes its PDF layout and nobody is responsible. Most mid-sized companies do not have the people for an AI operations team of their own.
Bitkom Research, KI in der deutschen Wirtschaft 2026, p. 12: 43% of AI users name a lack of staffMonitor and measure
Availability, quality and cost per case run into monitoring with alerts. If accuracy drops, we find the cause, test the fix and only then roll it out.
Model changes
Providers switch off older model versions on fixed dates, and the successor behaves differently. We test it on your cases before it decides anything for you.
The successor runs against the acceptance test set and new real cases. If it fails, we adjust and test again.
Larger changes and model switches only go live after your sign-off, at an announced time.
Every switch is in the change log with its test result. Otherwise nothing changes for your team.
14 October
Switched to the successor model, signed off by inside sales.done
What inviqon takes on
What your team takes on
Operation & improvement
Nothing in production yet? Start with the pilot and its acceptance criteria.
The next case is still open? The AI strategy ranks the candidates.
From the first call to a system in operation. Each phase stands on its own; none ties you to the next.
free
30 minutes. We work out where AI fits for you and whether an AI strategy is worth it.
€1,990
AI strategy for mid-sized companies: which of your processes suit AI, what the build costs and brings back, and which ones are not worth the effort.
from €5,000
AI pilot at a fixed price: one use case, fixed scope and acceptance criteria in writing up front. Built in your systems, tested with your data.
from 20% a year
Running AI systems: hosting in Germany, monitoring, model and prompt maintenance, ongoing measurement, and the next use case on the same foundation.
This page
Frequent questions
The questions that almost always come up.
From 20% of the project price per year, billed monthly and cancellable monthly. Example: at a project price of €30,000 that is €500 a month.
We test the successor against your acceptance test set first and only switch after your approval. Day to day, nothing changes except an entry in the change log.
From regular measurement against the same test set as at acceptance, extended with new real cases. You get the results in writing.
Yes. Code, data and documentation belong to you. We hand over to your IT or your IT service provider whenever you want.
In German data centres under a data processing agreement, or in your own environment.
Usually, because connections, permissions and data preparation are already in place. The price is fixed before it starts.
The next step
30 minutes, free and without obligation.
