← Back to community benchmarks

Qwen3.5-4B

M5 (8c) · 16 GB · 4bit · 2026-03-23
Performance
1k
tokens
1,071
PP tok/s
49.5
TG tok/s
957
TTFT (ms)
3.9
Peak mem (GB)
Hardware
Chip M5 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.19
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,071 49.5 3.9 GB current
4k 996.2 45.9 4.6 GB view
Batching Results
Batch Size TG tok/s Speedup
49.5 1.00×
89.7 1.81×
99.9 2.02×