← Back to community benchmarks

Qwen3.5-0.8B-OptiQ

M4 Pro (20c) · 64 GB · 4bit · 2026-03-15
Performance
8k
tokens
3,809
PP tok/s
216.7
TG tok/s
2151
TTFT (ms)
1.9
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.12
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,329 257.5 1.8 GB view
4k 3,660 238.8 1.9 GB view
8k 3,809 216.7 1.9 GB current
Batching Results
Batch Size TG tok/s Speedup
257.5 1.00×
451.5 1.75×
619.3 2.41×