← Back to community benchmarks

Qwen3.5-35B-A3BoQ4

M2 Max (30c) · 64 GB · 4bit · 2026-04-05
Performance
8k
tokens
549.2
PP tok/s
63.6
TG tok/s
14915
TTFT (ms)
21.1
Peak mem (GB)
Hardware
Chip M2 Max (30c)
Memory 64 GB
GPU Cores 30
Software
oMLX v0.3.4
macOS macOS 26.4
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 527.2 71.6 20.0 GB view
4k 601.1 68.8 20.7 GB view
8k 549.2 63.6 21.1 GB current
16k 516.4 52.7 21.7 GB view
32k 452.7 38.0 23.0 GB view