← Back to community benchmarks

Qwen3.6-35B-A3B

M5 Max (40c) · 128 GB · 8bit · 2026-04-25
Performance
32k
tokens
2,584
PP tok/s
64.8
TG tok/s
12683
TTFT (ms)
38.5
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 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,911 89.8 35.4 GB view
4k 2,353 88.5 36.2 GB view
16k 2,859 79.7 37.1 GB view
32k 2,584 64.8 38.5 GB current
Batching Results
Batch Size TG tok/s Speedup
89.8 1.00×
149.6 1.67×
219.6 2.45×
303.7 3.38×