← Back to community benchmarks

Qwen3.5-0.8B

M3 Pro (14c) · 36 GB · 8bit · 2026-04-19

Performance

1k

tokens

1,534

PP tok/s

110.6

TG tok/s

668

TTFT (ms)

1.6

Peak mem (GB)

Hardware

Chip M3 Pro (14c)

Memory 36 GB

GPU Cores 14

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	1,534	110.6	1.6 GB	current
4k	2,119	98.0	2.1 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	110.6	1.00×
2×	199.6	1.80×
4×	310.1	2.80×