← Back to community benchmarks

Qwen3.5-122B-A10B-RAM-48GB

M1 Max (32c) · 64 GB · 4bit · 2026-07-21

Performance

64k

tokens

104.1

PP tok/s

18.3

TG tok/s

629357

TTFT (ms)

49.2

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.5.2rc1

macOS macOS 26.5.2

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	177.5	28.4	44.8 GB	view
16k	162.3	24.4	47.5 GB	view
64k	104.1	18.3	49.2 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	28.4	1.00×
2×	30.9	1.09×
4×	61.8	2.18×