← Back to community benchmarks

Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2-abliterated-8bit

M2 Ultra (60c) · 64 GB · 8bit · 2026-06-04
Performance
4k
tokens
176.9
PP tok/s
19.7
TG tok/s
23159
TTFT (ms)
29.7
Peak mem (GB)
Hardware
Chip M2 Ultra (60c)
Memory 64 GB
GPU Cores 60
Software
oMLX v0.4.1
macOS macOS 26.2
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 173.2 20.1 28.3 GB view
4k 176.9 19.7 29.7 GB current
Batching Results
Batch Size TG tok/s Speedup
20.1 1.00×
32.9 1.64×
42.7 2.12×