← Back to community benchmarks

Qwen3.5-4B

M1 (7c) · 16 GB · 4bit · 2026-04-06
Performance
1k
tokens
45.1
PP tok/s
8.7
TG tok/s
22691
TTFT (ms)
3.3
Peak mem (GB)
Hardware
Chip M1 (7c)
Memory 16 GB
GPU Cores 7
Software
oMLX v0.3.4
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 45.1 8.7 3.3 GB current
4k 53.3 11.5 3.9 GB view
Batching Results
Batch Size TG tok/s Speedup
8.7 1.00×
14.3 1.64×
16.4 1.89×