Show HN: Qwen3-4B with the AI writing tics trained out

Explain Like I'm Normal

A developer fine-tuned the new Qwen3-4B model specifically to remove common 'AI-isms' and repetitive linguistic patterns often referred to as slop. By training on a dataset of high-quality human writing, the model aims to sound more natural and less like a corporate chatbot. This proof-of-concept shows that even small models can be significantly improved by targeted aesthetic fine-tuning.