Without options we are stuck with the defaults: https://docs.gpt4all.io/gpt4all_python.html#generation-parameters

def generate(
    prompt,
    max_tokens=200,
    temp=0.7,
    top_k=40,
    top_p=0.1,
    repeat_penalty=1.18,
    repeat_last_n=64,
    n_batch=8,
    n_predict=None,
    streaming=False,
):

That max_tokens=200 is particularly limiting.

Note that n_predict is a duplicate of max_tokens (for backwards compatibility) so I can ignore that one.