← Back to community benchmarks

Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled

M2 Max (30c) · 32 GB · 4bit · 2026-03-16
Performance
32k
tokens
264.6
PP tok/s
25.6
TG tok/s
123848
TTFT (ms)
7.8
Peak mem (GB)
Hardware
Chip M2 Max (30c)
Memory 32 GB
GPU Cores 30
Software
oMLX v0.2.13
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 301.4 60.3 5.8 GB view
4k 301.7 47.2 6.0 GB view
32k 264.6 25.6 7.8 GB current
Batching Results
Batch Size TG tok/s Speedup
60.3 1.00×
56.0 0.93×
63.1 1.05×