đ Excited to share our new paper, KTransformers-FineTune (KT-FT), developed in collaboration with the amazing
@llamafactory_ai team!
Fine-tune 671B MoE models with just 16GB GPU VRAM + CPU memory.
⥠2.5â21.2à faster than offloading baselines
đ§ Up to 128K-token contexts
đ§ CPU-GPU co-computation, beyond offloading
đ Paper: