← Back to community benchmarks

Qwen3.5-0.8B

M3 Max (30c) · 96 GB · 4bit · 2026-04-29

Performance

1k

tokens

2,432

PP tok/s

312.6

TG tok/s

421

TTFT (ms)

1.6

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 96 GB

GPU Cores 30

Software

oMLX v0.3.6

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,432	312.6	1.6 GB	current
4k	4,350	290.9	1.8 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	312.6	1.00×
2×	547.2	1.75×
4×	745.6	2.39×