Chantless: Taste-Grown Custom Models with 8B Beating Giants by 107.8%

tokenbender launched Chantless, a project for growing models aligned with personal taste; via full-model expert extraction, an 8B model beats a large model by 107.8% on function calling, culminating six months of research from early arithmetic and MLP experiments.

2026-10-14 ~ 2026-10-14 · 3 related posts

• Chantless launches: shaping models around your taste, chapter one — Apprentice — tokenbender · 2026-10-14
• Chantless specialist extraction: 8B model beats big model at function calling by 107.8% — tokenbender · 2026-10-14
• Chantless research trace: six months from arithmetic-MLP experiments to full-model specialists — tokenbender · 2026-10-14