← Back to community benchmarks

Qwen3.5-27B

M3 Max (40c) · 48 GB · 4bit · 2026-03-14
Performance
1k
tokens
216.2
PP tok/s
22.9
TG tok/s
4737
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.2.11
macOS macOS 26.3.1
Context 1,024
Batching Results
Batch Size TG tok/s Speedup
22.9 1.00×
39.1 1.71×
52.5 2.29×
55.9 2.44×