← Back to community benchmarks

Qwen3.5-122B-A10B

M3 Ultra (80c) · 256 GB · 8bit · 2026-03-23

Performance

4k

tokens

159.2

PP tok/s

15.4

TG tok/s

25733

TTFT (ms)

124.8

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 256 GB

GPU Cores 80

Software

oMLX v0.2.19

macOS macOS 15.7

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	142.1	16.3	123.3 GB	view
4k	159.2	15.4	124.8 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	16.3	1.00×
2×	18.3	1.12×
4×	20.3	1.25×