← Back to community benchmarks

Qwen3.6-35B-A3BVLM-oQ4

M1 Ultra (64c) · 64 GB · 4bit · 2026-04-24
Performance
16k
tokens
1,739
PP tok/s
55.7
TG tok/s
9423
TTFT (ms)
21.7
Peak mem (GB)
Hardware
Chip M1 Ultra (64c)
Memory 64 GB
GPU Cores 64
Software
oMLX v0.3.7rc2
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,073 71.1 19.9 GB view
4k 1,695 68.4 20.7 GB view
8k 1,791 65.0 21.0 GB view
16k 1,739 55.7 21.7 GB current
32k 1,462 38.5 23.0 GB view
64k 946.8 27.8 25.7 GB view
128k 512.3 16.4 31.1 GB view
195k 350.5 11.7 36.7 GB view