← Back to community benchmarks

Qwen3.6-27BoQ8

M1 Ultra (48c) · 128 GB · 8bit · 2026-05-11

Performance

4k

tokens

231.0

PP tok/s

18.0

TG tok/s

17730

TTFT (ms)

29.8

Peak mem (GB)

Hardware

Chip M1 Ultra (48c)

Memory 128 GB

GPU Cores 48

Software

oMLX v0.3.9.dev1

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	215.8	18.5	28.3 GB	view
4k	231.0	18.0	29.8 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	18.5	1.00×
2×	28.2	1.52×
4×	35.1	1.90×