Open-weight models can't write long stories: quality degrades or length undershoots

mrdrozdov · x · 2026-08-22

rishanthrajendh argues open-weight LLMs handle long-form storywriting poorly: either writing quality degrades as length grows, or they severely undershoot the requested length, often both. Frontier models fare better. The open question: can open-weight models be trained for long stories with limited data and compute?

Original post →

More from Models

Models channel →