← Back to community benchmarks

Qwen3.5-35B-A3B

M4 Pro (16c) · 48 GB · 4bit · 2026-03-09
Performance
16k
tokens
667.5
PP tok/s
61.6
TG tok/s
24544
TTFT (ms)
20.8
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 48 GB
GPU Cores 16
Software
oMLX v0.2.6
macOS macOS 26.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 705.2 85.9 20.1 GB view
4k 748.1 80.5 20.2 GB view
8k 723.2 73.9 20.4 GB view
16k 667.5 61.6 20.8 GB current
32k 573.9 48.8 21.6 GB view
Batching Results
Batch Size TG tok/s Speedup
85.9 1.00×
119.7 1.39×
141.1 1.64×