← Back to community benchmarks

Qwen3.5-27B

M5 Max (40c) · 128 GB · unknown · 2026-03-13
Performance
16k
tokens
521.6
PP tok/s
1.9
TG tok/s
31409
TTFT (ms)
59.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.9
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 333.0 1.9 57.8 GB view
4k 516.4 1.9 58.1 GB view
8k 542.0 2.0 58.7 GB view
16k 521.6 1.9 59.6 GB current
64k 359.4 1.8 69.2 GB view
Batching Results
Batch Size TG tok/s Speedup
1.9 1.00×
3.6 1.89×
7.1 3.74×
11.1 5.84×