← Back to community benchmarks

Qwopus3.5-122B-A10B-Kimi-K2.6-destill-healed-abliterated

M5 Max (40c) · 128 GB · 4bit · 2026-06-12

Performance

4k

tokens

1,530

PP tok/s

67.4

TG tok/s

2677

TTFT (ms)

66.9

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.4.4rc1

macOS macOS 26.5.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,038	69.6	65.6 GB	view
4k	1,530	67.4	66.9 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	69.6	1.00×
2×	97.0	1.39×
4×	119.8	1.72×