← Back to community benchmarks

Qwen3.5-35B-A3B-oQ4e

M3 Max (40c) · 128 GB · 4bit · 2026-04-18
Performance
16k
tokens
2,044
PP tok/s
34.9
TG tok/s
8016
TTFT (ms)
22.8
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
16k 2,044 34.9 22.8 GB current
32k 2,554 34.0 23.4 GB view