← Back to community benchmarks

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

M3 Ultra (60c) · 256 GB · 4bit · 2026-04-15
Performance
16k
tokens
311.5
PP tok/s
32.6
TG tok/s
52606
TTFT (ms)
19.1
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.5
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 305.0 38.6 15.8 GB view
4k 322.6 37.2 17.3 GB view
8k 319.5 35.5 17.9 GB view
16k 311.5 32.6 19.1 GB current
32k 292.8 27.6 21.6 GB view
Batching Results
Batch Size TG tok/s Speedup
38.6 1.00×
63.0 1.63×
72.2 1.87×
81.7 2.12×