← Back to community benchmarks

gemma-4-31B-it-Claude-Opus-Distilled8bit-mxfp8

M4 Pro (16c) · 48 GB · 8bit · 2026-04-18

Performance

1k

tokens

81.8

PP tok/s

7.2

TG tok/s

12514

TTFT (ms)

31.5

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.6

macOS macOS 26.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	81.8	7.2	31.5 GB	current
4k	81.1	6.9	33.3 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	7.2	1.00×
2×	11.3	1.57×
4×	28.6	3.97×