← Back to community benchmarks

qwen3.6-35b-a3b

M3 Ultra (60c) · 96 GB · 4bit · 2026-04-29
Performance
8k
tokens
2,126
PP tok/s
82.6
TG tok/s
3854
TTFT (ms)
21.6
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 96 GB
GPU Cores 60
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,576 360.8 21.8 GB view
4k 1,017 89.7 21.2 GB view
8k 2,126 82.6 21.6 GB current
16k 2,011 75.8 22.2 GB view
32k 1,741 49.0 23.5 GB view