← Back to community benchmarks

Qwen3.5-122B-A10B

M3 Ultra (60c) · 256 GB · 6bit · 2026-03-10

Performance

4k

tokens

697.0

PP tok/s

43.8

TG tok/s

5877

TTFT (ms)

95.1

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.2.6

macOS macOS 26.2

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	604.8	45.5	94.8 GB	view
4k	697.0	43.8	95.1 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	45.5	1.00×
2×	69.7	1.53×
4×	89.6	1.97×