← Back to community benchmarks

Qwen3.5-4B

M4 Max (32c) · 36 GB · 4bit · 2026-05-01
Performance
1k
tokens
1,064
PP tok/s
113.5
TG tok/s
962
TTFT (ms)
3.4
Peak mem (GB)
Hardware
Chip M4 Max (32c)
Memory 36 GB
GPU Cores 32
Software
oMLX v0.3.8
macOS macOS 15.3
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,064 113.5 3.4 GB current
4k 1,211 109.3 4.1 GB view