← Back to community benchmarks

Qwen3.6-27B

M3 Ultra (80c) · 256 GB · 8bit · 2026-04-26

Performance

4k

tokens

419.4

PP tok/s

20.2

TG tok/s

9765

TTFT (ms)

34.6

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 256 GB

GPU Cores 80

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	383.3	20.6	33.0 GB	view
4k	419.4	20.2	34.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	20.6	1.00×
2×	25.9	1.26×
4×	43.8	2.13×