← Back to community benchmarks

Qwen3.5-2B

M4 Max (40c) · 128 GB · 6bit · 2026-03-10
User commentOn battery.
Performance
32k
tokens
3,296
PP tok/s
136.9
TG tok/s
9941
TTFT (ms)
4.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
8k 3,832 188.7 3.2 GB view
16k 3,652 168.3 3.5 GB view
32k 3,296 136.9 4.1 GB current
64k 2,689 100.2 5.6 GB view