They released both GGUFs & custom llama.cpp fork today. 35B MOE in 7GB size which's good for Mobile & Edge devices(Also low memory systems). Up to 120 t/s on Consumer Laptop. GGUFs: https://huggingface.co/SyzygyResearch/Mach-1-Additive-35B-GGUF https://huggingface.co/SyzygyResearch/Mach-1-Additive-3
