← Back to community benchmarks

Qwen3.5-27B

M3 Ultra (80c) · 512 GB · 4bit · 2026-04-14

Performance

4k

tokens

414.2

PP tok/s

29.0

TG tok/s

9888

TTFT (ms)

34.4

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.5

macOS macOS 26.4

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	414.2	29.0	34.4 GB	current
64k	324.2	19.4	41.7 GB	view