← Back to community benchmarks

Qwen3.5-4B

M3 (8c) · 16 GB · 4bit · 2026-03-10
Performance
1k
tokens
275.9
PP tok/s
31.9
TG tok/s
3712
TTFT (ms)
3.9
Peak mem (GB)
Hardware
Chip M3 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 275.9 31.9 3.9 GB current
4k 273.3 30.5 4.2 GB view
Batching Results
Batch Size TG tok/s Speedup
31.9 1.00×
60.5 1.90×
56.0 1.76×