AnalysisAI ModelsAugust 6, 2026

Scotoma-2: Gemma 4 31B fine-tune with less slop, better writing

Community fine-tune of Gemma 4 31B targets repetitive sentence structures and 'slop' tics while improving prose quality; GGUF quantizations are available on HuggingFace.

1 source

Daily brief

Get tomorrow's AI brief in your inbox

More stories today

Open the live feed