← Back to community benchmarks

Qwen35-122B-A10BFP8-shared-expert

M2 Max (38c) · 96 GB · 4bit · 2026-03-20

Performance

32k

tokens

227.6

PP tok/s

18.8

TG tok/s

143985

TTFT (ms)

72.9

Peak mem (GB)

Hardware

Chip M2 Max (38c)

Memory 96 GB

GPU Cores 38

Software

oMLX v0.2.19

macOS macOS 26.3

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	230.1	34.2	67.3 GB	view
8k	256.4	29.8	69.4 GB	view
16k	246.7	24.3	70.6 GB	view
32k	227.6	18.8	72.9 GB	current
64k	192.9	11.4	77.8 GB	view