← Back to community benchmarks

Qwen3.5-VL-0.8B-8bitCRACK

M4 (10c) · 24 GB · 8bit · 2026-03-11

Performance

4k

tokens

1,964

PP tok/s

98.0

TG tok/s

2086

TTFT (ms)

1.9

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 24 GB

GPU Cores 10

Software

oMLX v0.2.7

macOS macOS 26.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,001	106.4	1.8 GB	view
4k	1,964	98.0	1.9 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	106.4	1.00×
2×	168.4	1.58×
4×	245.2	2.30×