← Back to community benchmarks

gpt-oss-20b-MXFP4-Q8

M2 Pro (19c) · 32 GB · 4bit · 2026-04-24

Performance

1k

tokens

297.7

PP tok/s

48.0

TG tok/s

3439

TTFT (ms)

11.8

Peak mem (GB)

Hardware

Chip M2 Pro (19c)

Memory 32 GB

GPU Cores 19

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	297.7	48.0	11.8 GB	current
4k	326.3	44.0	11.8 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	48.0	1.00×
2×	61.5	1.28×
4×	68.5	1.43×