How should I go about using ChatGPT's o3-mini-high?

[vermiculus]

If I understand correctly based on my research, o3-mini-high is 'just' o3-mini with a 'high' level of reasoning. o3-mini is available in gptel, but o3-mini-high is not -- and what's more, if I try to use o3-mini (which I do have access to on chatgpt.com, as well as o3-mini-high), I get a 404 'The model o3-mini does not exist or you do not have access to it.'

To use o3-mini-high, is there somehow a setting to configure on o3-mini (something that might need to be added to gptel-menu) or should this be set up as a different model altogether? Is my inability to access o3-mini from gptel a root cause here?

Sorry if this is a bit of a noob question; I've only started messing with this package this week. It's very impressive, though I've not thought much about different model use before now.

[karthink]

There appear to be two different problems. You can set the reasoning_effort parameter and use o3-mini as o3-mini-high, but I would sort this out first:

if I try to use o3-mini (which I do have access to on chatgpt.com, as well as o3-mini-high), I get a 404 'The model o3-mini does not exist or you do not have access to it.

Can you share your gptel configuration?