← Back to community benchmarks

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

M3 Pro (18c) · 36 GB · 4bit · 2026-05-01

Performance

1k

tokens

96.0

PP tok/s

8.7

TG tok/s

10671

TTFT (ms)

15.8

Peak mem (GB)

Hardware

Chip M3 Pro (18c)

Memory 36 GB

GPU Cores 18

Software

oMLX v0.3.8.dev3

macOS macOS 15.7.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	96.0	8.7	15.8 GB	current
4k	98.5	8.5	17.3 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	8.7	1.00×
2×	17.2	1.98×
4×	25.2	2.90×