← Back to community benchmarks

Qwen3.5-27B-Claude-4.6-Opus-Distilled

M2 Max (38c) · 64 GB · 6bit · 2026-03-16
Performance
195k
tokens
73.0
PP tok/s
3.6
TG tok/s
2740649
TTFT (ms)
44.0
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 64 GB
GPU Cores 38
Software
oMLX v0.2.13
macOS macOS 15.7.4
Context 200,000
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 109.5 15.3 22.0 GB view
4k 112.1 14.6 22.4 GB view
8k 111.5 13.9 22.8 GB view
16k 109.5 12.1 23.7 GB view
32k 105.3 10.1 25.5 GB view
64k 97.6 7.5 29.0 GB view
128k 84.0 5.0 36.4 GB view
195k 73.0 3.6 44.0 GB current
Batching Results
Batch Size TG tok/s Speedup
15.3 1.00×
20.8 1.36×
22.2 1.45×
20.5 1.34×