← Back to community benchmarks

Qwen3.5-122B-A10B3.7bit-VL

M4 Max (40c) · 128 GB · 3bit · 2026-04-30

Performance

195k

tokens

171.5

PP tok/s

12.2

TG tok/s

1166330

TTFT (ms)

83.0

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.8rc1

macOS macOS 26.4.1

Context 200,000