A Reddit user proposes that posts about model performance in r/LocalLLM should be required to include the quantization level in their title. They argue that omitting this detail leads to confusion and frustration when browsing the subreddit.