← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4e-fp16-mtp

M1 Max (32c) · 64 GB · 4bit · 2026-07-14
Performance
195k
tokens
252.5
PP tok/s
25.0
TG tok/s
792159
TTFT (ms)
24.9
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.5.1
macOS macOS 26.5.2
Context 200,000
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 875.7 63.8 20.1 GB view
4k 900.5 62.0 20.8 GB view
8k 809.4 60.7 20.9 GB view
16k 714.5 57.5 21.1 GB view
32k 607.5 51.5 21.6 GB view
64k 475.7 42.4 22.5 GB view
128k 328.9 32.2 23.4 GB view
195k 252.5 25.0 24.9 GB current
Batching Results
Batch Size TG tok/s Speedup
63.8 1.00×
95.3 1.49×
109.1 1.71×