← Back to community benchmarks

Qwen3.5-4B

M4 (10c) · 24 GB · 4bit · 2026-03-10
Performance
8k
tokens
334.2
PP tok/s
31.5
TG tok/s
24514
TTFT (ms)
4.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 355.7 35.5 3.9 GB view
4k 349.3 34.5 4.2 GB view
8k 334.2 31.5 4.6 GB current
Batching Results
Batch Size TG tok/s Speedup
35.5 1.00×
69.0 1.94×
68.7 1.94×
61.5 1.73×