← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4

M1 Pro (16c) · 32 GB · 4bit · 2026-04-17
Performance
1k
tokens
272.5
PP tok/s
43.9
TG tok/s
3758
TTFT (ms)
19.9
Peak mem (GB)
Hardware
Chip M1 Pro (16c)
Memory 32 GB
GPU Cores 16
Software
oMLX v0.3.6
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 272.5 43.9 19.9 GB current
4k 308.1 41.3 20.7 GB view
8k 304.9 37.6 21.1 GB view
16k 292.5 30.8 21.7 GB view
Batching Results
Batch Size TG tok/s Speedup
43.9 1.00×
56.4 1.28×
63.8 1.45×