← Back to community benchmarks

Qwen3.5-0.8B

M4 Pro (20c) · 48 GB · 8bit · 2026-04-29
Performance
16k
tokens
3,643
PP tok/s
165.2
TG tok/s
4498
TTFT (ms)
2.9
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,455 216.7 1.9 GB view
4k 3,719 200.5 2.1 GB view
8k 3,797 190.8 2.3 GB view
16k 3,643 165.2 2.9 GB current
Batching Results
Batch Size TG tok/s Speedup
216.7 1.00×
382.9 1.77×
532.9 2.46×
610.4 2.82×