← Back to community benchmarks

Qwen3.5-4B

M3 Max (40c) · 64 GB · 4bit · 2026-03-11
Performance
1k
tokens
1,309
PP tok/s
113.9
TG tok/s
782
TTFT (ms)
4.0
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.2.8
macOS macOS 15.7.4
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,309 113.9 4.0 GB current
4k 1,323 106.0 4.2 GB view
Batching Results
Batch Size TG tok/s Speedup
113.9 1.00×
195.3 1.71×
194.9 1.71×