← Back to community benchmarks

Qwen3.5-0.8B

M3 Max (40c) · 128 GB · 4bit · 2026-04-05

Performance

1k

tokens

2,640

PP tok/s

310.2

TG tok/s

388

TTFT (ms)

1.6

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.2

macOS macOS 26.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,640	310.2	1.6 GB	current
4k	2,930	293.7	1.9 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	310.2	1.00×
2×	616.1	1.99×
4×	585.6	1.89×