GPT-5-Nano for quick answer

Lukreaner

In the past, I found the Quick Answer feature to be awesome but also on the unreliable side. With the release of the GPT-5 models, it may be beneficial to switch the underlying model to GPT-5 Nano.

Using the Quick Answer feature but with improved output quality

Luis

Lukreaner thanks for the suggestion. The team consistently benchmarks new releases and selects the base model that performs best. I’ll mark the post as Done unless there’s anything else you’d like to ask (pls let follow up and I'll reopen)