This is completely separate from LoRA. This is how much stuff you can give it in the prompt. You can give it now whole chapters of books to summarize for example.
LoRA is for adapting the model to a certain model. It usually means you need to give it shorter prompts, but for book summarization, it wouldn't help.
LoRA probably does not affect the models biggest bottle neck. The attention mechanism. Original transformer was O(n^2d) where n is the query length and d was the cardinal of all the tokens
LoRA doesn’t change the context length. Also you can’t run LoRAs on GPT-4. So those are not relevant to each other.
The distinction becomes less meaningful if you can hold large numbers of tokens in attention, doesn’t it?