I write with Ollama on my own computer. Can long chapters overload my computer's memory?

From version 1.9.24, when "Ollama context window" in Settings is left on Automatic, BookyAI gives the model enough room to hold the chapter's instructions and its answer together, so long chapters are no longer cut short. It never makes that room bigger than your computer's memory can carry: about 8k tokens on a computer with 8 GB of memory, 16k with 16 GB and 32k with 32 GB. When a whole chapter does not fit at once, the model writes the rest in a continuation. On a computer with little memory, a smaller model and shorter chapters are the faster choice.