← Back to community benchmarks

Jundot-Qwen3.6-27B-oQ4-mtp

M5 Pro (16c) · 48 GB · 4bit · 2026-05-09

Performance

1k

tokens

358.6

PP tok/s

24.4

TG tok/s

2855

TTFT (ms)

16.8

Peak mem (GB)

Hardware

Chip M5 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.9.dev1

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	358.6	24.4	16.8 GB	current
4k	374.1	22.5	18.2 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	24.4	1.00×
2×	28.5	1.17×
4×	33.1	1.36×