You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Just tried using o1/o1-mini, faced the same problem.
Though my understanding of lua and nvim plugins ecosystem is limited, I believe that local configs for plugins override existing entries in plugins default tables, but still use default values for anything that is not present in the override.
After a quick look at the source code, it looks that max_tokens is a default parameter, hence is always present in request.
So far only o1-preview and o1-mini seem to be affected. But they are also still in beta and might change behavior in release version.
Still, though, is there any way to change request parameters based on model name? Or completely delete default config and send out your own parameters?
Firstly, thanks for this magnificient plugin.
I've been trying to use the o1-mini model but I get the following error:
I changed the configuration to add max_completion_tokens but it seems that the max_tokens is sent in the request.
anyone with similar issue?
The text was updated successfully, but these errors were encountered: