← Back to community benchmarks

gemma-4-31b-abliterated

M5 Pro (20c) · 64 GB · 4bit · 2026-04-05

Performance

1k

tokens

358.8

PP tok/s

14.5

TG tok/s

2854

TTFT (ms)

17.6

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.4

macOS macOS 26.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	358.8	14.5	17.6 GB	current
4k	355.6	13.1	19.4 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	14.5	1.00×
2×	24.1	1.66×
4×	56.8	3.92×