← Back to community benchmarks

gpt-oss-20b-MXFP4-Q8

M4 Max (32c) · 36 GB · 4bit · 2026-03-21

Performance

4k

tokens

915.2

PP tok/s

83.9

TG tok/s

4476

TTFT (ms)

11.5

Peak mem (GB)

Hardware

Chip M4 Max (32c)

Memory 36 GB

GPU Cores 32

Software

oMLX v0.2.12

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	820.8	91.8	11.4 GB	view
4k	915.2	83.9	11.5 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	91.8	1.00×
2×	124.6	1.36×
4×	155.1	1.69×