← Back to community benchmarks

Qwen3.6-35B-A3B

M5 Max (40c) · 128 GB · bf16 · 2026-04-18
Performance
16k
tokens
3,627
PP tok/s
64.3
TG tok/s
4518
TTFT (ms)
67.4
Peak mem (GB)
Hardware
Chip M5 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
1k 1,897 73.0 65.6 GB view
4k 3,455 70.2 66.4 GB view
8k 3,720 68.9 66.7 GB view
16k 3,627 64.3 67.4 GB current
32k 3,146 53.6 68.7 GB view
64k 2,314 39.3 71.3 GB view
128k 1,492 25.4 76.8 GB view
195k 1,086 18.3 82.5 GB view