← Back to community benchmarks

Qwen3.5-122B-A10B-5bit

M3 Max (40c) · 128 GB · 5bit · 2026-04-04

Performance

8k

tokens

478.5

PP tok/s

37.1

TG tok/s

17119

TTFT (ms)

81.9

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.2

macOS macOS 26.4

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	427.9	41.4	79.8 GB	view
4k	484.5	39.5	81.3 GB	view
8k	478.5	37.1	81.9 GB	current
16k	445.4	33.5	83.1 GB	view
32k	388.7	27.7	85.5 GB	view