← Back to community benchmarks

inferencerlabs-Qwen3.5-122B-A10B6.5bit

M4 Max (40c) · 128 GB · 6bit · 2026-03-30

Performance

1k

tokens

466.8

PP tok/s

43.3

TG tok/s

2194

TTFT (ms)

94.0

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.0rc1

macOS macOS 15.2

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	466.8	43.3	94.0 GB	current
4k	549.1	40.9	95.5 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	43.3	1.00×
2×	61.2	1.41×
4×	74.0	1.71×