The Engine and the Assistant
The champion answered the question correctly — and then produced three paragraphs of unrelated text. This is not a failure of the model. It is a failure of the layer that was never built on top of it.
The champion answered the question correctly — and then produced three paragraphs of unrelated text. This is not a failure of the model. It is a failure of the layer that was never built on top of it.
How we tamed gemma4:12b-mlx verbosity from 1537 tokens to 59 — and what it revealed about system prompt engineering.