← Back to community benchmarks

Qwen3.5-122B-A10B-Text-mxfp4

M2 Ultra (60c) · 128 GB · 4bit · 2026-03-27

Performance

1k

tokens

341.0

PP tok/s

47.6

TG tok/s

3003

TTFT (ms)

62.5

Peak mem (GB)

Hardware

Chip M2 Ultra (60c)

Memory 128 GB

GPU Cores 60

Software

oMLX v0.2.22

macOS macOS 26.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	341.0	47.6	62.5 GB	current
4k	398.5	44.8	63.9 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	47.6	1.00×
2×	68.3	1.43×
4×	81.6	1.71×