← Back to community benchmarks

Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated

M5 Pro (20c) · 64 GB · 8bit · 2026-05-04
Performance
32k
tokens
1,725
PP tok/s
54.0
TG tok/s
18994
TTFT (ms)
38.9
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,360 65.8 35.4 GB view
4k 2,093 64.6 36.2 GB view
8k 2,163 62.0 36.7 GB view
16k 2,060 60.0 37.4 GB view
32k 1,725 54.0 38.9 GB current
64k 1,281 44.0 41.9 GB view
Batching Results
Batch Size TG tok/s Speedup
65.8 1.00×
106.3 1.62×
142.3 2.16×
285.0 4.33×