๐ 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.
โก 2.5โ21.2ร faster than offloading baselines
๐ง Up to 128K-token contexts
๐ง CPU-GPU co-computation, beyond offloading
๐ Paper: