← Back to community benchmarks

Qwen3.5-122B-A10B

M2 Ultra (76c) · 128 GB · 6bit · 2026-03-22

Performance

4k

tokens

440.5

PP tok/s

38.5

TG tok/s

9299

TTFT (ms)

95.1

Peak mem (GB)

Hardware

Chip M2 Ultra (76c)

Memory 128 GB

GPU Cores 76

Software

oMLX v0.2.12

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	404.6	40.3	94.8 GB	view
4k	440.5	38.5	95.1 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	40.3	1.00×
2×	64.2	1.59×
4×	82.8	2.05×