yavarb/moefit

Run Qwen3.8-Flash-Next on Apple Silicon Macs that cannot hold the full MoE in RAM — expert paging, pinned hot-set, exact routing-replay sidecar (repeat KEEP ~+40%).

3 best points · 0 comments · 1 Hacker News post · 3 GitHub stars · View on GitHub

Hacker News discussions

• Moefit: Fit Qwen 3.8 Flash on your Mac with limited ram (3 points, 0 comments, by ybathaee, 2026-10-07)