Tag

Qwen3

2 stories

Sebastian Raschka's free 11-hour course trains a reasoning model from Qwen3 0.6B (magazine.sebastianraschka.com)

The course improves a pre-trained Qwen3 0.6B model rather than training from scratch, runs early sections on a CPU and ships code in a GitHub repository.

Commenters split between readers of his LLM book who loved it and skeptics who asked who the course is for.

🚀Tech ★ 564 · 💬 81 · yesterday · #AI#programming#PyTorch#Qwen3#Sebastian Raschka

Not just a headline — the discussion unpacked

The top HN threads: facts from the discussions, both sides, insider voices. Email, Telegram, or RSS.

Show HN: Qwen3-4B fine-tuned to drop em dashes cut its 'slop score' from 121 to 12.5 (100toolkit.com)

Erik Markus Kannike released qwen3-4b-deslop-v4, which needs no system prompt. On 40 held-out prompts, em dashes fell from 20.5 to zero and 68% of answers landed in human-writing range.

Commenters found it still reads as AI, and one asked why anyone would want that.