← Back to community benchmarks

Qwen3.5-122B-A10B-heretic-v2-2.34bit-msq

M2 Ultra (76c) · 128 GB · 4bit · 2026-04-16

Performance

1k

tokens

360.5

PP tok/s

45.6

TG tok/s

2841

TTFT (ms)

41.5

Peak mem (GB)

Hardware

Chip M2 Ultra (76c)

Memory 128 GB

GPU Cores 76

Software

oMLX v0.3.6

macOS macOS 26.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	360.5	45.6	41.5 GB	current
4k	466.8	43.4	42.8 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	45.6	1.00×
2×	75.9	1.66×
4×	100.2	2.20×