Get Zlio on your home screenOne tap, full screen, like an app.
Open Models · first seen 14 November 2026, 19:12 UTC
tandem.cpp runs DeepSeek-V4.1 on 64GB PCs, adds AMD ROCm
tandem.cpp is a patch set for llama.cpp that runs MoE models too large for VRAM by keeping hot experts on the GPU and offloading the rest to CPU. Update adds DeepSeek-V4.1-Flash support on a 64GB PC with a 16GB card, plus AMD ROCm support.