Qwen3.8 125B (6B active) Flash-Next
on M5 Max 128GB · mlx-serve · 1,048,576 ctx
Use cases
long-contextcoding
Summary
User reports about 40 t/s on prose and 75 t/s on coding at roughly 760k context.
Setup uses 8-bit dense layers and 4-bit expert layers, with peak memory around 117 GB and a requirement of iogpu.wired_limit_mb=120000. Model weights are at ddalcu/Qwen3.8-Flash-Next-MLX-Serve-mixed-4-8bit.