← Back to community benchmarks

gpt-oss-120b-MXFP4-Q8

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-14

Performance

4k

tokens

717.9

PP tok/s

61.3

TG tok/s

5706

TTFT (ms)

59.3

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.2.10

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	595.6	76.3	59.2 GB	view
4k	717.9	61.3	59.3 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	76.3	1.00×
2×	110.2	1.44×
4×	150.1	1.97×