fuxiaopang
local.ai/fuxiaopang
Claimed August 2026
Referred by V2EX
My machines
Mac Studio M4 Max
ApproximateApple · 36 GB
Approximate options
Choose what matters most
Best fit
Qwen3.6 35B A3B · UD-Q3_K_M
Best balance of speed and intelligence.
Intelligence66.4Speed12.2 tasks/hrSize16.6 GBMore intelligent
Qwen3.6 27B OptiQ · 4BIT
~8% more intelligence · ~151% slower.
Intelligence71.9Speed4.9 tasks/hrSize19.7 GB
Faster
Holo 3.1 35B A3B · Q4_K_M
~41% faster · retains ~78% intelligence.
Intelligence52.0Speed20.6 tasks/hrSize21.3 GBApproximate for 36 GB VRAM. Results may vary.