← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4

M5 Max (40c) · 128 GB · 4bit · 2026-05-04
Performance
4k
tokens
3,866
PP tok/s
121.7
TG tok/s
1060
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,636 124.5 19.9 GB view
4k 3,866 121.7 20.7 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 124.5 1.00×
2× 193.7 1.56×
4× 280.2 2.25×
8× 367.5 2.95×