← Back to community benchmarks

Qwen3.5-35B-A3B

M4 Pro (16c) · 48 GB · 6bit · 2026-03-09
Performance
4k
tokens
721.5
PP tok/s
62.2
TG tok/s
5677
TTFT (ms)
28.3
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 48 GB
GPU Cores 16
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 675.5 65.5 28.1 GB view
4k 721.5 62.2 28.3 GB current
Batching Results
Batch Size TG tok/s Speedup
65.5 1.00×
84.8 1.29×
95.8 1.46×