← Back to community benchmarks

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

M5 Pro (20c) · 64 GB · 6bit · 2026-03-25
Performance
64k
tokens
327.2
PP tok/s
9.3
TG tok/s
200307
TTFT (ms)
33.1
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.20
macOS macOS 26.4
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 394.8 13.0 22.1 GB view
4k 439.6 12.7 23.6 GB view
32k 389.1 10.9 28.1 GB view
64k 327.2 9.3 33.1 GB current
Batching Results
Batch Size TG tok/s Speedup
13.0 1.00×
23.3 1.79×
28.5 2.19×