← Back to community benchmarks

Ternary-Bonsai-27B

M5 Pro (20c) · 64 GB · 2bit · 2026-07-23

Performance

32k

tokens

142.0

PP tok/s

23.4

TG tok/s

230813

TTFT (ms)

14.7

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.5.3

macOS macOS 26.5.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	482.2	30.7	8.9 GB	view
32k	142.0	23.4	14.7 GB	current
64k	130.2	18.9	19.7 GB	view