← Back to community benchmarks

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

M5 Pro (20c) · 48 GB · 6bit · 2026-03-17
Performance
16k
tokens
421.6
PP tok/s
11.7
TG tok/s
38864
TTFT (ms)
23.7
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.2.14
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 395.5 12.9 22.0 GB view
4k 439.5 12.7 22.4 GB view
8k 436.8 12.4 22.8 GB view
16k 421.6 11.7 23.7 GB current
32k 390.1 10.6 25.5 GB view
64k 328.2 8.9 29.0 GB view
Batching Results
Batch Size TG tok/s Speedup
12.9 1.00×
22.7 1.76×
27.9 2.16×
28.9 2.24×