← Back to community benchmarks

Jackrong-Qwen3.5-9B-Claude-Reasoning-abliterated-mxfp8

M4 Pro (16c) · 24 GB · 8bit · 2026-03-12

Performance

4k

tokens

320.8

PP tok/s

26.4

TG tok/s

12769

TTFT (ms)

9.8

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 24 GB

GPU Cores 16

Software

oMLX v0.2.9

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	310.5	26.9	9.6 GB	view
4k	320.8	26.4	9.8 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	26.9	1.00×
2×	31.7	1.18×
4×	34.2	1.27×