← Back to community benchmarks

Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-8bit-abliterated

M2 Max (38c) · 96 GB · 8bit · 2026-03-10
Performance
4k
tokens
156.5
PP tok/s
11.1
TG tok/s
26166
TTFT (ms)
32.1
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 96 GB
GPU Cores 38
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 154.0 11.5 31.3 GB view
4k 156.5 11.1 32.1 GB current
Batching Results
Batch Size TG tok/s Speedup
11.5 1.00×
20.8 1.81×
27.5 2.39×