← Back to community benchmarks

Huihui-Qwen3.5-35B-A3B-Claude-4.6-Opus-abliterated

M5 Pro (16c) · 48 GB · 4bit · 2026-04-16

Performance

4k

tokens

1,854

PP tok/s

92.3

TG tok/s

2209

TTFT (ms)

20.2

Peak mem (GB)

Hardware

Chip M5 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.2.13

macOS macOS 26.4

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,714	97.9	20.1 GB	view
4k	1,854	92.3	20.2 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	97.9	1.00×
2×	149.8	1.53×
4×	189.5	1.94×