← Back to community benchmarks

Qwen3.5-35B-A3Blm

M1 Pro (14c) · 32 GB · 4bit · 2026-03-10

Performance

8k

tokens

260.9

PP tok/s

36.4

TG tok/s

31394

TTFT (ms)

18.6

Peak mem (GB)

Hardware

Chip M1 Pro (14c)

Memory 32 GB

GPU Cores 14

Software

oMLX v0.2.7

macOS macOS 15.6

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	239.2	43.7	18.3 GB	view
4k	263.6	41.5	18.4 GB	view
8k	260.9	36.4	18.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	43.7	1.00×
2×	52.0	1.19×
4×	58.1	1.33×