← Back to community benchmarks

gpt-oss-20b-MXFP4-Q8

M4 Pro (16c) · 24 GB · 4bit · 2026-03-10

Performance

1k

tokens

430.6

PP tok/s

66.6

TG tok/s

2378

TTFT (ms)

11.4

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 24 GB

GPU Cores 16

Software

oMLX v0.2.7

macOS macOS 15.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	430.6	66.6	11.4 GB	current
4k	498.0	54.5	11.5 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	66.6	1.00×
2×	89.2	1.34×
4×	97.8	1.47×