neopolita / Niwaki-Flash-Next-23GB-v4.0-gguf

GGUF build of Niwaki-Flash-Next-19GB-v4.0-mlx — Qwen3.8-Flash-Next at 119B total / ~5.6B active parameters — for llama.cpp on CUDA, Vulkan, Metal and CPU. One 40.2 GB file whose GPU part is 21.1 GiB: the 15.6 GiB n-gram embedding table stays in CPU memory, read from the file a…