← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4

M3 Max (30c) · 36 GB · 4bit · 2026-04-16
Performance
64k
tokens
601.3
PP tok/s
30.1
TG tok/s
108993
TTFT (ms)
25.7
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 36 GB
GPU Cores 30
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 631.6 66.0 19.9 GB view
4k 821.7 66.7 20.7 GB view
64k 601.3 30.1 25.7 GB current
Batching Results
Batch Size TG tok/s Speedup
66.0 1.00×
120.4 1.82×
148.2 2.25×