← Back to community benchmarks

Step-3.5-Flash-oQ3

M5 Max (40c) · 128 GB · 3bit · 2026-04-05

Performance

8k

tokens

873.3

PP tok/s

45.2

TG tok/s

9382

TTFT (ms)

83.6

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.2

macOS macOS 26.4

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	593.2	52.4	83.3 GB	view
4k	878.8	48.6	83.5 GB	view
8k	873.3	45.2	83.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	52.4	1.00×
2×	77.3	1.48×
4×	97.9	1.87×