← Back to community benchmarks

Qwen3.5-27B

M1 Max (24c) · 32 GB · 4bit · 2026-03-11
Performance
16k
tokens
65.5
PP tok/s
11.4
TG tok/s
250183
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 32 GB
GPU Cores 24
Software
oMLX v0.2.7
macOS macOS 15.6
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 67.2 15.5 16.6 GB view
4k 67.3 14.5 17.0 GB view
8k 66.7 13.4 17.4 GB view
16k 65.5 11.4 18.3 GB current
Batching Results
Batch Size TG tok/s Speedup
15.5 1.00×
17.0 1.10×
34.1 2.20×
68.1 4.39×