AnalysisAI ModelsAugust 6, 2026

Scotoma-2: Gemma4 fine-tune with less slop and better writing

Scotoma-2 is a community fine-tune of the Gemma 4 31B IT model, released on Hugging Face with GGUF quantizations. The creators say it targets model tics like repetitive sentence structures while improving writing quality.

1 source

Daily brief

Get tomorrow's AI brief in your inbox

More stories today

Open the live feed