← Back to community benchmarks

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

M3 Ultra (80c) · 512 GB · 6bit · 2026-03-08
Performance
16k
tokens
392.6
PP tok/s
22.4
TG tok/s
41733
TTFT (ms)
23.7
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.6
macOS macOS 15.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 375.2 26.7 22.0 GB view
4k 401.1 25.4 22.4 GB view
8k 400.8 24.4 22.8 GB view
16k 392.6 22.4 23.7 GB current
32k 370.0 19.5 25.5 GB view
Batching Results
Batch Size TG tok/s Speedup
26.7 1.00×
45.8 1.72×
55.5 2.08×
60.8 2.28×