← Back to community benchmarks

Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled

M4 Max (32c) · 36 GB · 4bit · 2026-03-15

Performance

1k

tokens

580.0

PP tok/s

64.7

TG tok/s

1766

TTFT (ms)

5.8

Peak mem (GB)

Hardware

Chip M4 Max (32c)

Memory 36 GB

GPU Cores 32

Software

oMLX v0.2.12

macOS macOS 15.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	580.0	64.7	5.8 GB	current
4k	640.0	63.6	6.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	64.7	1.00×
2×	120.5	1.86×
4×	152.8	2.36×